Discussion:
Xaprb: 50 things to know before migrating Oracle to MySQL
(too old to reply)
Ariz Jacinto
2009-03-13 06:14:52 UTC
Permalink
FTA:
http://www.xaprb.com/blog/2009/03/13/50-things-to-know-before-migrating-oracle-to-mysql/

"[...]

1. Subqueries are poorly optimized.
2. Complex queries are a weak point.
3. The query executioner (aka query optimizer / planner) is less
sophisticated.
4. Performance tuning and metrics capabilities are limited.
5. There is limited ability to audit.
6. Security is unsophisticated, even crude. There are no groups or roles,
no ability to deny a privilege (you can only grant privileges). A user who
logs in with the same username and password from different network addresses
may be treated as a completely separate user. There is no built-in
encryption comparable to Oracle.
7. Authentication is built-in. There is no LDAP, Active Directory, or
other external authentication capability.
8. Clustering is not what you think it is.
9. Stored procedures and triggers are limited.
10. Vertical scalability is poor.
11. There is zero MPP support.
12. SMP is supported, but MySQL doesn’t scale well to more than 4 or 8
cores/CPUs.
13. There is no fractional-second storage type for times, dates, or
intervals.
14. The language used to write stored procedures, triggers, scheduled
events, and stored functions is very limited.
15. There is no roll-back recovery. There is only roll-forward recovery.
16. There is no support for snapshots.
17. There is no support for database links. There is something called the
Federated storage engine that acts as a relay by passing queries along to a
table on a remote server, but it is crude and buggy.
18. Data integrity checking is very weak, and even basic integrity
constraints cannot always be enforced.
19. There are very few optimizer hints to tune query execution plans.
20. There is only one type of join plan: nested-loop. There are no
sort-merge joins or hash joins.
21. Most queries can use only a single index per table; some multi-index
query plans exist in certain cases, but the cost is usually underestimated
by the query optimizer, and they are often slower than a table scan.
22. There are no bitmap indexes. Each storage engine supports different
types of indexes. Most engines support B-Tree indexes.
23. There are fewer and less sophisticated tools for administration.
24. There is no IDE and debugger that approaches the level of
sophistication you may be accustomed to. You’ll probably be writing your
stored procedures in a text editor and debugging them by adding statements
that insert rows into a table called debug_log.
25. Each table can have a different storage backend (”storage engine”).
26. Each storage engine can have widely varying behavior, features, and
properties.
27. Foreign keys are not supported in most storage engines.
28. The default storage engine is non-transactional and corrupts easily.
29. Oracle owns InnoDB, the most advanced and popular storage engine.
30. Certain types of execution plans are only supported in some storage
engines. Certain types of COUNT() queries execute instantly in some storage
engines and slowly in others.
31. Execution plans are not cached globally, only per-connection.
32. Full-text search is limited and only available for non-transactional
storage backends. Ditto for GIS/spatial types and queries.
33. There are no resource controls. A completely unprivileged user can
effortlessly run the server out of memory and crash it, or use up all CPU
resources.
34. There are no integrated or add-on business intelligence, OLAP cube,
etc packages.
35. There is nothing analogous to Grid Control.
36. There is nothing even remotely like RAC. If you are asking “How do I
build RAC with MySQL,” you are asking the wrong question.
37. There are no user-defined types or domains.
38. The number of joins per query is limited to 61.
39. MySQL supports a smaller subset of SQL syntax. There are no recursive
queries, common table expressions, or windowing functions. There are a few
extensions to SQL that are somewhat analogous to MERGE and similar features,
but are very simplistic in comparison.
40. There are no functional columns (e.g. a column whose value is
calculated as an expression).
41. You cannot create an index on an expression, you can only index
columns.
42. There are no materialized views.
43. The statistics vary between storage engines and regardless of the
storage engine, are limited to simple cardinality and rows-in-a-range. In
other words, statistics on data distribution are limited. There is not much
control over updating of statistics.
44. There is no built-in promotion or failover mechanism.
45. Replication is asynchronous and has many limitations and edge cases.
For example, it is single-threaded, so a powerful slave can find it hard to
replicate fast enough to keep up with a less powerful master.
46. Cluster is not what you think it is. Maybe I already said that, but
it bears repeating.
47. The data dictionary (INFORMATION_SCHEMA) is limited and very slow (it
can easily crash a busy server).
48. There is no online ALTER TABLE.
49. There are no sequences.
50. DDL such as ALTER TABLE or CREATE TABLE is non-transactional. It
commits open transactions and cannot be rolled back or crash-recovered.
Schema is stored in the filesystem independently of the storage engine."
Orlando Andico
2009-03-13 06:52:35 UTC
Permalink
This is a refreshing change from all the "Migrating from Oracle to
MySQL" spam that Sun fills my inbox with.

(the price to pay when registering in order to download the JDK)

There's way much more I can add such as...

51. There is no support for multimedia data types such as WAV, MOV,
QT, MPEG4, AIFF-C, etc. etc. and there is no trans-coding support

52. There is no support for important image types for document and
medical imaging such as JPEG2000, DICOM, etc. etc., there is no
resizing/annotation/EXIF metadata editing support, and there is no
image querying ("look for the image with this color" or "look for the
image with a texture similar to this one")

53. There is no built-in JVM and there is no way to build stored procs
in a non-bloody way. You cannot run Quercus inside MySQL and thus have
a PHP interpreter resident inside your database.

54. There is no support for spatial functions such as shortest-path,
nearest-neighbor, etc.
Post by Ariz Jacinto
http://www.xaprb.com/blog/2009/03/13/50-things-to-know-before-migrating-oracle-to-mysql/
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Tito Mari Francis Escaño
2009-03-13 07:22:58 UTC
Permalink
How does PostgreSQL match-up to the features required from Oracle? I
recall EnterpriseDB once claimed they can directly migrate data from
Oracle to EnterpriseDB, a rebranded PostgreSQL with some extensions.
Maybe it's time to reconsider a not-so-popular but feature-wise more
appropriate free and open source database server?
Post by Ariz Jacinto
http://www.xaprb.com/blog/2009/03/13/50-things-to-know-before-migrating-oracle-to-mysql/
"[...]
Subqueries are poorly optimized.
Complex queries are a weak point.
The query executioner (aka query optimizer / planner) is less sophisticated.
Performance tuning and metrics capabilities are limited.
There is limited ability to audit.
Security is unsophisticated, even crude. There are no groups or roles, no
ability to deny a privilege (you can only grant privileges). A user who logs
in with the same username and password from different network addresses may
be treated as a completely separate user. There is no built-in encryption
comparable to Oracle.
Authentication is built-in. There is no LDAP, Active Directory, or other
external authentication capability.
Clustering is not what you think it is.
Stored procedures and triggers are limited.
Vertical scalability is poor.
There is zero MPP support.
SMP is supported, but MySQL doesn’t scale well to more than 4 or 8
cores/CPUs.
There is no fractional-second storage type for times, dates, or intervals.
The language used to write stored procedures, triggers, scheduled events,
and stored functions is very limited.
There is no roll-back recovery. There is only roll-forward recovery.
There is no support for snapshots.
There is no support for database links. There is something called the
Federated storage engine that acts as a relay by passing queries along to a
table on a remote server, but it is crude and buggy.
Data integrity checking is very weak, and even basic integrity constraints
cannot always be enforced.
There are very few optimizer hints to tune query execution plans.
There is only one type of join plan: nested-loop. There are no sort-merge
joins or hash joins.
Most queries can use only a single index per table; some multi-index query
plans exist in certain cases, but the cost is usually underestimated by the
query optimizer, and they are often slower than a table scan.
There are no bitmap indexes. Each storage engine supports different types of
indexes. Most engines support B-Tree indexes.
There are fewer and less sophisticated tools for administration.
There is no IDE and debugger that approaches the level of sophistication you
may be accustomed to. You’ll probably be writing your stored procedures in a
text editor and debugging them by adding statements that insert rows into a
table called debug_log.
Each table can have a different storage backend (”storage engine”).
Each storage engine can have widely varying behavior, features, and
properties.
Foreign keys are not supported in most storage engines.
The default storage engine is non-transactional and corrupts easily.
Oracle owns InnoDB, the most advanced and popular storage engine.
Certain types of execution plans are only supported in some storage engines.
Certain types of COUNT() queries execute instantly in some storage engines
and slowly in others.
Execution plans are not cached globally, only per-connection.
Full-text search is limited and only available for non-transactional storage
backends. Ditto for GIS/spatial types and queries.
There are no resource controls. A completely unprivileged user can
effortlessly run the server out of memory and crash it, or use up all CPU
resources.
There are no integrated or add-on business intelligence, OLAP cube, etc
packages.
There is nothing analogous to Grid Control.
There is nothing even remotely like RAC. If you are asking “How do I build
RAC with MySQL,” you are asking the wrong question.
There are no user-defined types or domains.
The number of joins per query is limited to 61.
MySQL supports a smaller subset of SQL syntax. There are no recursive
queries, common table expressions, or windowing functions. There are a few
extensions to SQL that are somewhat analogous to MERGE and similar features,
but are very simplistic in comparison.
There are no functional columns (e.g. a column whose value is calculated as
an expression).
You cannot create an index on an expression, you can only index columns.
There are no materialized views.
The statistics vary between storage engines and regardless of the storage
engine, are limited to simple cardinality and rows-in-a-range. In other
words, statistics on data distribution are limited. There is not much
control over updating of statistics.
There is no built-in promotion or failover mechanism.
Replication is asynchronous and has many limitations and edge cases. For
example, it is single-threaded, so a powerful slave can find it hard to
replicate fast enough to keep up with a less powerful master.
Cluster is not what you think it is. Maybe I already said that, but it bears
repeating.
The data dictionary (INFORMATION_SCHEMA) is limited and very slow (it can
easily crash a busy server).
There is no online ALTER TABLE.
There are no sequences.
DDL such as ALTER TABLE or CREATE TABLE is non-transactional. It commits
open transactions and cannot be rolled back or crash-recovered. Schema is
stored in the filesystem independently of the storage engine."
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
--
Tito Mari Francis H. Escaño
Computer Engineer and Free Software Proponent
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://arc
joebert jacaba
2009-03-13 13:36:35 UTC
Permalink
I'm not impressed unless Oracle is already free and open source.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Jerome Gotangco
2009-03-13 13:51:48 UTC
Permalink
Post by joebert jacaba
I'm not impressed unless Oracle is already free and open source.
Does free and open source make a product compelling automatically?
--
Jerome G.

Blog: http://blog.gotangco.com
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
joebert jacaba
2009-03-13 13:56:42 UTC
Permalink
Post by Jerome Gotangco
Post by joebert jacaba
I'm not impressed unless Oracle is already free and open source.
Does free and open source make a product compelling automatically?
No. The same reason a commercial one doesn't.
Post by Jerome Gotangco
--
Jerome G.
Blog: http://blog.gotangco.com
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Jerome Gotangco
2009-03-13 13:57:52 UTC
Permalink
On Fri, Mar 13, 2009 at 3:22 PM, Tito Mari Francis Escaño
Post by Tito Mari Francis Escaño
How does PostgreSQL match-up to the features required from Oracle? I
recall EnterpriseDB once claimed they can directly migrate data from
Oracle to EnterpriseDB, a rebranded PostgreSQL with some extensions.
Maybe it's time to reconsider a not-so-popular but feature-wise more
appropriate free and open source database server?
I think its unfair to compare Oracle 10 or 11g to Postgres. There are
a lot of verticals that Oracle cover that Postgres probably doesn't
because of market demand. A more appropriate comparison would probably
be OracleXE to PostgreSQL.
--
Jerome G.

Blog: http://blog.gotangco.com
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
joebert jacaba
2009-03-13 14:12:31 UTC
Permalink
Post by Jerome Gotangco
I think its unfair to compare Oracle 10 or 11g to Postgres. There are
a lot of verticals that Oracle cover that Postgres probably doesn't
because of market demand. A more appropriate comparison would probably
be OracleXE to PostgreSQL.
And is it fair to compare it to MySQL?
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Jerome Gotangco
2009-03-13 14:16:38 UTC
Permalink
Post by joebert jacaba
Post by Jerome Gotangco
I think its unfair to compare Oracle 10 or 11g to Postgres. There are
a lot of verticals that Oracle cover that Postgres probably doesn't
because of market demand. A more appropriate comparison would probably
be OracleXE to PostgreSQL.
And is it fair to compare it to MySQL?
We can probably make it head to head with Postgres, MySQL, OracleXE
and MS SQL Server :-)

At work we have a heavy bias to Postgres for a lot of reasons,
primarily the way it does backups along with PITR which MySQL isn't
really that great with bin logs, etc. but we do use MySQL because of
market demand, and in some places, IBM DB2 Express-C
--
Jerome G.

Blog: http://blog.gotangco.com
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
joebert jacaba
2009-03-13 14:26:37 UTC
Permalink
Use what works for you, your client, your brain, your pocket, your
advocacy. If I can use 1% or Oracle features then I'll be happy. Most
applications are just shi++y PHP+MySQL anyway.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-13 16:59:44 UTC
Permalink
Sun actually positions MySQL as an Oracle replacement in some cases,
and you ought to check out their support contract pricing.. the
comparison is more relevant than you guys think.
Post by joebert jacaba
Post by Jerome Gotangco
I think its unfair to compare Oracle 10 or 11g to Postgres. There are
a lot of verticals that Oracle cover that Postgres probably doesn't
because of market demand. A more appropriate comparison would probably
be OracleXE to PostgreSQL.
And is it fair to compare it to MySQL?
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Winelfred G. Pasamba
2009-03-19 04:05:23 UTC
Permalink
Is there are possible market state where Oracle will zero the
license-to-use price and only charge for support?
Post by Orlando Andico
Sun actually positions MySQL as an Oracle replacement in some cases,
and you ought to check out their support contract pricing.. the
comparison is more relevant than you guys think.
Post by joebert jacaba
Post by Jerome Gotangco
I think its unfair to compare Oracle 10 or 11g to Postgres. There are
a lot of verticals that Oracle cover that Postgres probably doesn't
because of market demand. A more appropriate comparison would probably
be OracleXE to PostgreSQL.
And is it fair to compare it to MySQL?
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-19 07:09:08 UTC
Permalink
i cannot comment on possible future directions but that is akin to
asking if Microsoft will make Office and the OS free and just charge
for support.

Proprietary pricing is not exclusive to Oracle, and the huge closed
source market proves that it is a valid business model and that
clients will pay premium prices for enterprise functionality.

It's like a bicycle and a Benz S500. If you just want to get from A to
B and don't mind sweating, go for the bicycle (or do open source).
Feature sets are so far apart that a real technical comparison is
meaningless.
Post by Winelfred G. Pasamba
Is there are possible market state where Oracle will zero the
license-to-use price and only charge for support?
Post by Orlando Andico
Sun actually positions MySQL as an Oracle replacement in some cases,
and you ought to check out their support contract pricing.. the
comparison is more relevant than you guys think.
Post by joebert jacaba
Post by Jerome Gotangco
I think its unfair to compare Oracle 10 or 11g to Postgres. There are
a lot of verticals that Oracle cover that Postgres probably doesn't
because of market demand. A more appropriate comparison would probably
be OracleXE to PostgreSQL.
And is it fair to compare it to MySQL?
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
joebert jacaba
2009-03-19 15:49:52 UTC
Permalink
Post by Orlando Andico
It's like a bicycle and a Benz S500. If you just want to get from A to
B and don't mind sweating, go for the bicycle (or do open source).
Feature sets are so far apart that a real technical comparison is
meaningless.
A bicycle doesn't emit gases harmful to the environment. It is not
just sweat but a whole lot of health benefits as well. It is much much
cheaper and can go to any point A to any point B as well. Maybe a lot
slower but with more practice can go faster. Much faster on traffic
jams as well and can pass single tracks and more roads not passable or
not economical for a car, much more a Benz, can pass.

It's a philosophy not end (or money) justfies the means thing.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Paolo Falcone
2009-03-19 18:08:42 UTC
Permalink
A bicycle won't get you faster than a Benz S500 if you're needed 15
minutes from now and you're about to travel on a highway. Similarly,
using MySQL to compute your 3-dimensional report (like a sales report
that spans multiple years) via OLTP is just begging for pain.

It's naive (and sometimes fatal) for a business not to use the right
tool for the job. Comparing MySQL and Oracle is like comparing apples
and oranges. MySQL has a long way to go (way too long) to even match
Oracle's current capabilities, given that Oracle has a decade+ head
start, and they do know what they're doing in the RDBMS realm, among
others (not to mention scaling horizontally and vertically).

But if you don't need Oracle's bloat and MySQL's current features
suits you fine, or perhaps you have special needs that Oracle's DB
won't fit, then go use MySQL.
Post by joebert jacaba
Post by Orlando Andico
It's like a bicycle and a Benz S500. If you just want to get from A to
B and don't mind sweating, go for the bicycle (or do open source).
Feature sets are so far apart that a real technical comparison is
meaningless.
A bicycle doesn't emit gases harmful to the environment. It is not
just sweat but a whole lot of health benefits as well. It is much much
cheaper and can go to any point A to any point B as well. Maybe a lot
slower but with more practice can go faster. Much faster on traffic
jams as well and can pass single tracks and more roads not passable or
not economical for a car, much more a Benz, can pass.
It's a philosophy not end (or money) justfies the means thing.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
joebert jacaba
2009-03-19 21:56:02 UTC
Permalink
Post by Paolo Falcone
A bicycle won't get you faster than a Benz S500 if you're needed 15
minutes from now and you're about to travel on a highway.
good luck on the streets of manila and on a rush hour. so it really
depends on the situation.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Paolo Falcone
2009-03-20 02:09:43 UTC
Permalink
A bicycle won't help either :D Dodging with the trucks and buses,
you're better off in a car.

Seriously, it's really situational. There's no 'one solution that fits
all'. For most technologies, depending on your niche, it's the 'one
solution that fits most' that wins the bid.
Post by joebert jacaba
Post by Paolo Falcone
A bicycle won't get you faster than a Benz S500 if you're needed 15
minutes from now and you're about to travel on a highway.
good luck on the streets of manila and on a rush hour. so it really
depends on the situation.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Jerome Gotangco
2009-03-20 02:12:08 UTC
Permalink
Post by Paolo Falcone
Seriously, it's really situational. There's no 'one solution that fits
all'. For most technologies, depending on your niche, it's the 'one
solution that fits most' that wins the bid.
At the lowest possible price that can yield high value as well
--
Jerome G.

Blog: http://blog.gotangco.com
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Drexx Laggui [personal]
2009-03-20 02:22:16 UTC
Permalink
20Mar2009 (UTC +8)

Dudes, the analogies are getting ridiculous... that instead of
enlightening non-DBA's like me, it's just confusing me even more :P

On the lighter side, how's this for a crazy analogy? Roller-blading in
EDSA during rush hour!

That's exactly what I'm thinking when back in early-1990's, I coded a
DBMS using Turbo Pascal 4 on a 486DX with 4MB RAM, for a college
"instructor" of mine so he can complete *his* thesis. From nothing to
a beta version in 2 weeks. Talk about hard work that eventually
enabled me to go where I needed to go. Ha ha ha!!!
Post by Jerome Gotangco
Post by Paolo Falcone
Seriously, it's really situational. There's no 'one solution that fits
all'. For most technologies, depending on your niche, it's the 'one
solution that fits most' that wins the bid.
At the lowest possible price that can yield high value as well
Drexx Laggui -- CISA, CISSP, CFE Associate, ISO27001 LA, CCSI, CSA
http://www.laggui.com ( Singapore / Manila / California )
Computer forensics; Penetration testing; QMS & ISMS developers; K-Transfer
PGP fingerprint = 6E62 A089 E3EA 1B93 BFB4 8363 FFEC 3976 FF31 8A4E
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Rowel Atienza
2009-03-20 02:49:58 UTC
Permalink
Yeah. Gets confusing. The analogies presented do not make sense.

The bottom line is what ever suits your intended app. Boils down to
good software engineering decision.

We even use databases as small as sqlite with 100k records on some of
our apps where MySQL is overkill and Oracle will not sense at all. In
those apps, sqlite works elegantly.

rowel
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-20 02:59:46 UTC
Permalink
My "bicycle" analogy was because there's a pervasive attitude in this
community to "roll your own" and look with contempt at closed-source
offerings. Heck, I have that attitude myself....

However, just like a Benz S500, there are people whose needs and
requirements are beyond our comprehension (I've never seen the inside
of a Benz S500, so I'm sure it has features which I don't know I
need!!! but which some people would pay big bucks for).

It's exactly the same with software, like, databases. I did use
Oracle8 and 8i in my previous job, but my perception of Oracle was
that it was this huge complicated very reliable database, but it was
just a database.

To someone who's used only MySQL, that is a realistic mindset. But
it's extremely far from the truth. Oracle DB has features you'd never
dream existed. Because you don't know that they exist, you think
Oracle is useless and thus a legitimate comparison to MySQL is
warranted.

But people say who want to archive X-Ray images in DICOM format (with
metadata), and they need it encrypted with 3rd-party-validated
encryption to meet HIPAA standards... you can't do that with MySQL.
But a hospital with that need will pay almost any price to get it.

The Xarpb fellow (Ronald) who wrote the original comparison is, for
reference, the author of "High Performance MySQL." Yes, he is a MySQL
community contributor and author. He is also ex-Oracle.

FWIW.
Post by Rowel Atienza
Yeah. Gets confusing. The analogies presented do not make sense.
The bottom line is what ever suits your intended app. Boils down to
good software engineering decision.
We even use databases as small as sqlite with 100k records on some of
our apps where MySQL is overkill and Oracle will not sense at all. In
those apps, sqlite works elegantly.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Rafael Sevilla
2009-03-20 03:21:17 UTC
Permalink
On Fri, 20 Mar 2009 10:59:46 +0800
Post by Orlando Andico
However, just like a Benz S500, there are people whose needs and
requirements are beyond our comprehension (I've never seen the inside
of a Benz S500, so I'm sure it has features which I don't know I
need!!! but which some people would pay big bucks for).
A better analogy then might be to say that Oracle is akin to an 18
wheeler truck. Certainly there are many situations where such a huge and
complex vehicle is the only thing that can get the job done, but there
are just many situations where such a thing is massive overkill and
it may in fact not be agile enough to cut the mustard.
--
You are still innocent until proven guilty. What has changed is
what they do to innocent people.
http://stormwyrm.blogspot.com
John Peter Loh
2009-03-20 03:01:02 UTC
Permalink
I was thinking street luge at EDSA would be crazier.

I agree on the right tool for the right job. MySQL works fine on small
to mid-sized websites. It's free, lightweight and pretty much does the
job.

It's a different story when you move to an enterprise app. I had my
fair share of problems on scaling with MySQL. There are solutions but
it's different when you're battling against deadlines.

Only one wish for Oracle: Support for more distros. Probably too much to ask.
Post by Drexx Laggui [personal]
20Mar2009 (UTC +8)
Dudes, the analogies are getting ridiculous... that instead of
enlightening non-DBA's like me, it's just confusing me even more :P
On the lighter side, how's this for a crazy analogy? Roller-blading in
EDSA during rush hour!
That's exactly what I'm thinking when back in early-1990's, I coded a
DBMS using Turbo Pascal 4 on a 486DX with 4MB RAM, for a college
"instructor" of mine so he can complete *his* thesis. From nothing to
a beta version in 2 weeks. Talk about hard work that eventually
enabled me to go where I needed to go. Ha ha ha!!!
Post by Jerome Gotangco
Post by Paolo Falcone
Seriously, it's really situational. There's no 'one solution that fits
all'. For most technologies, depending on your niche, it's the 'one
solution that fits most' that wins the bid.
At the lowest possible price that can yield high value as well
Drexx Laggui  -- CISA, CISSP, CFE Associate, ISO27001 LA, CCSI, CSA
http://www.laggui.com  ( Singapore / Manila / California )
Computer forensics; Penetration testing; QMS & ISMS developers; K-Transfer
PGP fingerprint = 6E62 A089 E3EA 1B93 BFB4  8363 FFEC 3976 FF31 8A4E
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-20 03:08:04 UTC
Permalink
On Fri, Mar 20, 2009 at 11:01 AM, John Peter Loh <j.peterloh-***@public.gmane.org> wrote:
..
Post by John Peter Loh
Only one wish for Oracle: Support for more distros. Probably too much to ask.
RHEL and OEL and SuSE are not enough??

Lemme guess, more Ubuntu users.

Oracle supports the distros which are popular enough and
revenue-generating. I don't think there are enough Ubuntu-using
companies which can pay $48K/CPU for DB EE.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
John Peter Loh
2009-03-20 03:11:42 UTC
Permalink
I was thinking ArchLinux (and FreeBSD). He he. Like I said, too much to ask.
Post by Orlando Andico
..
Post by John Peter Loh
Only one wish for Oracle: Support for more distros. Probably too much to ask.
RHEL and OEL and SuSE are not enough??
Lemme guess, more Ubuntu users.
Oracle supports the distros which are popular enough and
revenue-generating. I don't think there are enough Ubuntu-using
companies which can pay $48K/CPU for DB EE.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Rafael Sevilla
2009-03-20 03:43:47 UTC
Permalink
On Fri, 20 Mar 2009 11:11:42 +0800
Post by John Peter Loh
I was thinking ArchLinux (and FreeBSD). He he. Like I said, too much to ask.
In practice what you really get when you buy an Oracle license isn't so
much the right to simply use the software as the right to ask for
support from Oracle when things go wrong, and in mission-critical
environments, enterprises tend to want to know that they'll have
someone to blame and fall back on to fix it immediately when that
inevitably happens. That means that the operating system must also be
covered by a similar support contract in case it's operating system
issues that are causing problems. Can you buy an enterprise support
contract for ArchLinux or FreeBSD? Didn't think so. Perhaps someday
Canonical might make overtures to Oracle to begin supporting Oracle on
Ubuntu Server Edition, which was, after all, the same process Red Hat
and Novell did to get Oracle support for their distributions, but so
far it would seem that Canonical isn't making moves in this direction.

Someone has to ask Oracle to support their software on some platform.
They aren't going to simply pick up some random platform not backed
by anyone and get Oracle to run on it just because, since doing so
represents a serious investment on their part without clear
possibilities for return.
--
You are still innocent until proven guilty. What has changed is
what they do to innocent people.
http://stormwyrm.blogspot.com
joebert jacaba
2009-03-20 14:14:35 UTC
Permalink
Post by Paolo Falcone
A bicycle won't help either :D Dodging with the trucks and buses,
you're better off in a car.
i just biked EDSA today from philcoa to magallanes. on my way home i
passed C5. its not hard as long as you practice often and care for the
environment. :)
Post by Paolo Falcone
Seriously, it's really situational. There's no 'one solution that fits
all'. For most technologies, depending on your niche, it's the 'one
solution that fits most' that wins the bid.
that's the point exactly.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-20 18:27:33 UTC
Permalink
I like the bicycle analogy.

But just me, I would not take a bike if I could take a Benz. I think
this is a sentiment shared by most of the folks here....

That's the thing... yes you can do anything with <pick your choice of
technology>. But you may not want to...

"One solution that fits the most" I heartily agree with. The trouble
is that some advocate MySQL as the be-all, end-all Oracle replacement.
That's crazy talk.

MySQL often mentions 80% of the functionality of Oracle -- what's more
accurate is 80% of the base functionality. If that's all you need,
then great! but if your business depends on the 20%, not so great...
my past experience with MySQL InnoDB storage engine was that
concurrency is not that good, you can deadlock the database. MyISAM
was far, far worse. That's pretty basic functionality I think.



On Fri, Mar 20, 2009 at 10:14 PM, joebert jacaba <joebertj-***@public.gmane.org> wrote:
..
Post by joebert jacaba
i just biked EDSA today from philcoa to magallanes. on my way home i
passed C5. its not hard as long as you practice often and care for the
environment. :)
Post by Paolo Falcone
Seriously, it's really situational. There's no 'one solution that fits
all'. For most technologies, depending on your niche, it's the 'one
solution that fits most' that wins the bid.
that's the point exactly.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
joebert jacaba
2009-03-22 14:32:53 UTC
Permalink
Post by Orlando Andico
I like the bicycle analogy.
i would say it is not well thought of as others pointed out some much
better alternatives.
Post by Orlando Andico
But just me, I would not take a bike if I could take a Benz. I think
this is a sentiment shared by most of the folks here....
if you have just said a generic car then i think it is much better. or
a motorcycle perhaps.

you may be bias to oracle but i admit i am bias to mysql and the
bicycle. i ride my bike at least 100km per week. i bike for clean air
and more. to me "philosophy" weighs much more than any features you
may list. since it can be contested the clear choice to me is still
mysql unless it ceases to be open source. same as ill ditch my bike
for a car if it will no longer emit gases harmful to the environment.

some things are much more expensive than others not because of
functionality but just because of the "bling."
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-22 15:27:16 UTC
Permalink
Joebert,

I don't doubt your reasoning at all. I say "bicycle" because generally
(actually with Open Source in general) a lot more work is required to
get a certain level of functionality and availability working.

But as you say -- the bicycle gets you there, sometimes later rather
than sooner. I cannot bike 100km a week unlike you or my brother -- so
for me the bicycle is simply out of the question. Same as many
enterprises, they cannot afford long development cycles, or a lot of
work to ensure high availability, etc.

But as I mentioned in my prior email, if I was starting out building
something, I'd probably build it with MySQL (or PostgreSQL) to avoid
large up-front costs. It's easy to migrate from MySQL to Oracle if
warranted (but the other way around is not so easy..)

Although I do not think the feature set of Oracle is bling -- some
people need what they need. Things like ensuring HIPAA compliance of
the entire database so you can ensure privacy of medical data stored
in it: these things cost money to develop and certify. I would not
consider that bling, if I needed it. Or for example Transparent Data
Encryption and Database Vault, which are required for PCI-DSS
compliance.
Post by joebert jacaba
Post by Orlando Andico
I like the bicycle analogy.
i would say it is not well thought of as others pointed out some much
better alternatives.
Post by Orlando Andico
But just me, I would not take a bike if I could take a Benz. I think
this is a sentiment shared by most of the folks here....
if you have just said a generic car then i think it is much better. or
a motorcycle perhaps.
you may be bias to oracle but i admit i am bias to mysql and the
bicycle. i ride my bike at least 100km per week. i bike for clean air
and more. to me "philosophy" weighs much more than any features you
may list. since it can be contested the clear choice to me is still
mysql unless it ceases to be open source. same as ill ditch my bike
for a car if it will no longer emit gases harmful to the environment.
some things are much more expensive than others not because of
functionality but just because of the "bling."
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
joebert jacaba
2009-03-24 14:22:12 UTC
Permalink
orly,

i have a high respect for you being a pioneer in linux no doubt about it.
Post by Orlando Andico
Joebert,
I don't doubt your reasoning at all. I say "bicycle" because generally
(actually with Open Source in general) a lot more work is required to
get a certain level of functionality and availability working.
tell me how easy it is to produce a sum of money to buy an oracle license.
Post by Orlando Andico
But as you say -- the bicycle gets you there, sometimes later rather
than sooner. I cannot bike 100km a week unlike you or my brother -- so
for me the bicycle is simply out of the question. Same as many
enterprises, they cannot afford long development cycles, or a lot of
work to ensure high availability, etc.
because they are thinking short term. they can invest on open source
now. it doesnt mean if you can't do it now you drop the idea right
away. i started biking catching my breath after 1km of pedaling. after
6 months i can bike 100km a day 3 days straight. it takes dedication
and practice.
Post by Orlando Andico
But as I mentioned in my prior email, if I was starting out building
something, I'd probably build it with MySQL (or PostgreSQL) to avoid
large up-front costs.
im not impressed. me ill pick mysql because in my own little way i
could contribute back to the community and not because i can save on
cost. that's the philosophy im talking about.
Post by Orlando Andico
It's easy to migrate from MySQL to Oracle if
warranted (but the other way around is not so easy..)
because oracle doesn't want to. i think mysql will be glad if oracle
share to them what they need to make that easy. but that is trade
secret where you draw cash right?
Post by Orlando Andico
Although I do not think the feature set of Oracle is bling -- some
people need what they need. Things like ensuring HIPAA compliance of
the entire database so you can ensure privacy of medical data stored
in it: these things cost money to develop and certify. I would not
consider that bling, if I needed it. Or for example Transparent Data
Encryption and Database Vault, which are required for PCI-DSS
compliance.
compliance is a business thing. it is clearly a bling. with a
relatively unlimited amount of money heck it is easy to get
compliance. ooxml? im not certified in anything myself because im
confident in my skills.

joebert
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-24 14:32:06 UTC
Permalink
On Tue, Mar 24, 2009 at 10:22 PM, joebert jacaba <joebertj-***@public.gmane.org> wrote:
..
Post by joebert jacaba
compliance is a business thing. it is clearly a bling. with a
relatively unlimited amount of money heck it is easy to get
compliance. ooxml? im not certified in anything myself because im
confident in my skills.
It's not about skill. It's about the confidence level of
line-of-business users. Your confidence in your skillset is irrelevant
to the highest echelons of large companies.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
joebert jacaba
2009-03-27 15:14:55 UTC
Permalink
Post by Orlando Andico
It's not about skill. It's about the confidence level of
line-of-business users. Your confidence in your skillset is irrelevant
to the highest echelons of large companies.
i would like to change where they base their confidence level. it
should not be on how much money a company has to buy compliance but on
the philosophy behind the solution. in my own little way.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
andrelst
2009-03-28 16:35:39 UTC
Permalink
Post by joebert jacaba
Post by Orlando Andico
It's not about skill. It's about the confidence level of
line-of-business users. Your confidence in your skillset is irrelevant
to the highest echelons of large companies.
i would like to change where they base their confidence level. it
should not be on how much money a company has to buy compliance but on
the philosophy behind the solution. in my own little way.
Joebert,

It can be done, but in the real world it is very hard to do. I used to
be a young grasshopper and idealistic. Have become a pragmatic guy...
dedicated logic and common sense, and suffer the consequence whether
it's good or bad.

But there is still some idealism left. Give it another 10 years, you
become a skeptic and shout "Get off my lawn!" :)

Companies don't buy software because of philosophy of "Open Source" or
it's free, they buy it based on support, scalability and features.
They don't ask, "is this FAX Machine hardware or software open
source?". It's like attempting to use Zimbra because it's Open Source
to replace MS Exchange... and your boss is scratching his head why,
when 25,000 of the company uses MS Exchange and they are okay with it.

But you sometimes gamble to thrust Open Source software because you
know it's the right tool for the right job. Like when I pushed to
include rsync on one of the key software to manage almost a thousand
physical servers on Solaris, AIX and HP-UX everyday on a bank. In
retrospect, It was a bit preposterous, putting my job on the line. But
hey, you know it works you pushed for it slowly but surely.

regards,
Andre
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
joebert jacaba
2009-03-30 15:08:54 UTC
Permalink
Post by andrelst
It can be done, but in the real world it is very hard to do. I used to
be a young grasshopper and idealistic. Have become a pragmatic guy...
dedicated logic and common sense, and suffer the consequence whether
it's good or bad.
even if it is very very hard and can be done then lets do it.
Post by andrelst
But there is still some idealism left. Give it another 10 years, you
become a skeptic and shout "Get off my lawn!" :)
that happens if you didn't get very rich by then. i agree my mileage
maybe shorter say 5 years. but before that happens then ill stay this
way.
Post by andrelst
Companies don't buy software because of philosophy of "Open Source" or
it's free, they buy it based on support, scalability and features.
most companies are really there to earn. take for example a plastic or
a cigarette company. they should know that what they produce will be
detrimental to the environment or health but they do it anyway.
Post by andrelst
They don't ask, "is this FAX Machine hardware or software open
source?".
they should start asking. on the hardware part just be sure spare
parts are not proprietary.
Post by andrelst
It's like attempting to use Zimbra because it's Open Source
to replace MS Exchange... and your boss is scratching his head why,
when 25,000 of the company uses MS Exchange and they are okay with it.
you are a good candidate to replace your boss.
Post by andrelst
But you sometimes gamble to thrust Open Source software because you
know it's the right tool for the right job. Like when I pushed to
include rsync on one of the key software to manage almost a thousand
physical servers on Solaris, AIX and HP-UX everyday on a bank. In
retrospect, It was a bit preposterous, putting my job on the line. But
hey, you know it works you pushed for it slowly but surely.
wow! galing mo.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-30 17:10:48 UTC
Permalink
Post by joebert jacaba
Post by andrelst
It can be done, but in the real world it is very hard to do. I used to
be a young grasshopper and idealistic. Have become a pragmatic guy...
dedicated logic and common sense, and suffer the consequence whether
it's good or bad.
even if it is very very hard and can be done then lets do it.
Now here's why this statement makes no sense:

We always pay people to do things we don't like doing.

For example, changing your motor oil is a dirty, potentially dangerous
job (hot used motor oil! carcinogenic!) but one time I bought the
motor oil, oil filter, filter wrench etc. I ended up paying a mechanic
300 pesos to do it.

I also did my own clutch cylinder replacement. I got it done!!! and
got bleeding hands in the process.

People always will pay someone else to do something that's too hard,
or too unpleasant for them to do.

Heck, we pay the government (indirectly, via taxes) to cart away our
trash. We pay Malabanan to shovel our shit away.

Following Joebert's reasoning, we should all be shoveling our own shit..
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-30 17:45:19 UTC
Permalink
My last tract for the day..

First, a comment on RMS' ethos, which is probably shared by Joebert and others.

RMS' most important point is: Freedom is the most important thing. We
should never compromise on any software solution which requires us to
surrender our freedom.

Agreed?

But here's the thing... most if not all people are willing to give up
some freedom in exchange for other considerations, which may include
money, well-being, safety, convenience...

The UK s the most surveilled society in the world, but the Britons
aren't complaining because they feel safer.

Singapore has one of the most punitive criminal codes in the world.
Smoke on the subway? $500 fine! and few people dare to question the
government. Singapore really can be called an authoritarian state. But
look at all the Pinoys flocking there... they'd rather be heavily
taxed on cigarettes and liquor, and not be able to litter or denounce
the government, in exchange for an orderly existence.

In any organized society, you by definition give up some freedom. So
RMS' whole argument about the supreme sanctity of freedom, is to me
over-rated.


Now for a slightly more technical point.... anyone here ever heard or
work with MaxDB?

Long long ago, in a galaxy far far away, SAP licensed the IP rights
for Software AG's database Adabas-D. They forked the code and produced
their own OLTP, robust database called... SAP DB.

It's even certified to run SAP!

Now SAP apparently got tired of maintaining their own database (not
their core business) so they open-sourced it and let MySQL sell it as
MaxDB. Can you believe this? truly open-source, enterprise-class
database!!

So how many people on this list / Oracle haters actually USE MaxDB?

I'm waiting.....

The inconvenient truth (which I experienced first hand when I learned
that the MaxDB "community" mostly consisted of a guy named Tilo
Heinrich who works for SAP in Berlin) is that nobody from the Free
software community turned out to help SAP make MaxDB a truly
ass-kicking or should I say Oracle-kicking database.

Don't believe me? do a google search for "andico maxdb" -- I actually
tried to get this stuff working, but when even the installer doesn't
work and the only guy who will help you comes from SAP -- I gave up.
So much for the Free software community.

Like I said w.r.t. Mikael Ronstrom, there are certain software
disciplines that your average Free software hacker can't comprehend.
So nobody helps out on these projects. I submit that database
technology is one of those disciplines (and I'm talking 21st century
database, not Oracle version 6, circa-1989 technology).


More major examples: OpenOffice and Java. Both open-source. But both
of them would wilt without Sun, as the vast majority of the engineers
working on them are Sun employees. Schwartz may have pulled a dumb
stunt, open-sourcing Sun's crown jewels without getting a veritable
Mongolian Horde of Free software hackers to take over developing these
systems.

Bottom-line, developing complex software costs money. And time and
again, the Free software community has shown that as you go up the
value chain, the capability of the community to contribute (CCC!)
becomes constrained :-)
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
joebert jacaba
2009-03-30 20:50:24 UTC
Permalink
Orly,

As if you speak, you speak for the rest of the world. You have your
opinions and i have mine. I maybe wrong and you are right. I am not
saying I will stay this way and not be Orly in the future. As long im
still happy and contented with my life i will be dogmatic/idealistic
whatever you may call it.
To you because you would rather pay than to go through the troubles.
Post by Orlando Andico
We always pay people to do things we don't like doing.
that's you. most people the poor ones every minute of their life does
things they dont like doing.
Post by Orlando Andico
For example, changing your motor oil is a dirty, potentially dangerous
job (hot used motor oil! carcinogenic!) but one time I bought the
motor oil, oil filter, filter wrench etc. I ended up paying a mechanic
300 pesos to do it.
if you were not determined to finish what you have started then you
have just paid the mechanic right from the start.
Post by Orlando Andico
I also did my own clutch cylinder replacement. I got it done!!! and
got bleeding hands in the process.
galeng! at least you have done it. the secret is to enjoy while doing it.

i build my bikes from ground up. of course i have hurt my hands but i
get satisfaction from it rather than hate. heck i enjoy butingting my
bikes more than riding it.
Post by Orlando Andico
People always will pay someone else to do something that's too hard,
or too unpleasant for them to do.
so do they pay you because its hard or unpleasant? i work and they pay
me and i think my job is petiks and enjoyable.
Post by Orlando Andico
Heck, we pay the government (indirectly, via taxes) to cart away our
trash.
too general example. you can segregate your waste, minimize use of
plastics, recycle!
Post by Orlando Andico
We pay Malabanan to shovel our shit away.
Following Joebert's reasoning, we should all be shoveling our own shit..
great deduction skills here.

just enjoy your life. i can't tell you what to do with it. im just
telling you what i do with mine and hope others would do the same.

Joebert
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Johann Vincent Paul Tagle
2009-03-25 01:56:16 UTC
Permalink
Hi all,
Post by joebert jacaba
Post by Orlando Andico
But as you say -- the bicycle gets you there, sometimes later rather
than sooner. I cannot bike 100km a week unlike you or my brother -- so
for me the bicycle is simply out of the question. Same as many
enterprises, they cannot afford long development cycles, or a lot of
work to ensure high availability, etc.
because they are thinking short term. they can invest on open source
now. it doesnt mean if you can't do it now you drop the idea right
away. i started biking catching my breath after 1km of pedaling. after
6 months i can bike 100km a day 3 days straight. it takes dedication
and practice.
I don't think they are thinking short term but thinking what makes sense
with regard to their business. They need such a feature now, they can pay
for it, so they go with what's available and can be reliably supported now.
One of your objectives in your biking must be to get healthy, and there's
really no shortcut to it but dedication and practice. However, in business,
there are many ways to accomplish something, and what you choose should be
what makes sense to you. If you are making say 25M per year on a product
using an Oracle database, the initial 2M investment and the succeeding 0.5M
yearly support cost is nothing. Even if in the long run the open source
alternative is a lot cheaper, they won't be willing to make the change,
because what they have simply works for them.

Johann
--
Web application and database consulting: www.sagadasolutions.com
Orlando Andico
2009-03-25 02:13:03 UTC
Permalink
At the risk of diverging this thread again...

I gave some more thought to Joebert's assertions. There are a couple
things to consider:

1) compliance is not bling. For example if you want to interconnect
with MasterCard's network, you NEED to be PCI-DSS compliance. This
compliance is audited by say SGS.

2) If you want to store financial data from US-domiciled entities, you
need to comply with Sarbenes-Oxley regulations. Again this is the law.

3) If you want to store health-care data, you need to comply with HIPAA.

Now I am not saying these things cannot be done with open-source. But
it's a lot harder. For example PCI-DSS requires role separation
between the data owner and the DBA. Can this be done with MySQL? I
think not. This is to say nothing of the root user! also, data on the
disk and over the wire must be encrypted. True you can use SSL or an
encrypting file system for this, but it's not a "validated" solution.

Financial services and health care is big business. If you were a bank
with a large turnover, would you go with a Free solution, or one
backed by a billion-dollar company?

Another assertion I'll make (which may be controversial) is that in
the long run, Free solutions are not necessarily cheaper than Closed
ones. While it is true that the license cost is negligible, ongoing
support costs are comparable and may be more -- if for example you
cannot Go To Market in time because you're still DIY'ing your
solution.

Free solutions have their place -- if you have a crack tech team or
you're a startup. But most large companies don't have the luxury of
hiring the Joeberts and Foolers of this world -- they have to make do
with grunts and their personnel churn is high. In such a case, you
need to have a single neck to strangle, and that's the large Closed
Source vendor.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Winelfred G. Pasamba
2009-03-26 02:05:57 UTC
Permalink
so, if i have a system that runs with mysql or postgresql, what are
the business benefits of having it also run with oracle?
does it make sense to raise the price if the client opt to use oracle?

or do those clients who can pay for an oracle license belong to the
market where SAP, Oracle, & others dominate?

why is oracle focusing on linux? does this mean big companies are
starting to not-want-to-pay other platforms?

of course we know that the percent of product revenue is going down in
comparison to support revenue, except maybe for microsoft and adobe.*

is it still true that IBM is still the largest database vendor?*


*The Business of Software: What Every Manager, Programmer, and
Entrepreneur Must Know to Thrive and Survive in Good Times and Bad.
Michael A. Cusumano
Post by Orlando Andico
At the risk of diverging this thread again...
I gave some more thought to Joebert's assertions. There are a couple
1) compliance is not bling. For example if you want to interconnect
with MasterCard's network, you NEED to be PCI-DSS compliance. This
compliance is audited by say SGS.
2) If you want to store financial data from US-domiciled entities, you
need to comply with Sarbenes-Oxley regulations. Again this is the law.
3) If you want to store health-care data, you need to comply with HIPAA.
Now I am not saying these things cannot be done with open-source. But
it's a lot harder. For example PCI-DSS requires role separation
between the data owner and the DBA. Can this be done with MySQL? I
think not. This is to say nothing of the root user! also, data on the
disk and over the wire must be encrypted. True you can use SSL or an
encrypting file system for this, but it's not a "validated" solution.
Financial services and health care is big business. If you were a bank
with a large turnover, would you go with a Free solution, or one
backed by a billion-dollar company?
Another assertion I'll make (which may be controversial) is that in
the long run, Free solutions are not necessarily cheaper than Closed
ones. While it is true that the license cost is negligible, ongoing
support costs are comparable and may be more -- if for example you
cannot Go To Market in time because you're still DIY'ing your
solution.
Free solutions have their place -- if you have a crack tech team or
you're a startup. But most large companies don't have the luxury of
hiring the Joeberts and Foolers of this world -- they have to make do
with grunts and their personnel churn is high. In such a case, you
need to have a single neck to strangle, and that's the large Closed
Source vendor.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
--
Discover the meaning of the Middle East Conflict...
http://www.discoveries08.org/schedule


Winelfred G. Pasamba
Adventist University of the Philippines Online Information Systems
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-26 02:33:00 UTC
Permalink
On Thu, Mar 26, 2009 at 10:05 AM, Winelfred G. Pasamba
Post by Winelfred G. Pasamba
so, if i have a system that runs with mysql or postgresql, what are
the business benefits of having it also run with oracle?
does it make sense to raise the price if the client opt to use oracle?
It depends.

Ramcar for example is running on Postgresql. But if they want to use
Demantra demand-planning software, they may have to get Oracle
licenses.

Generally people moving from Free software to closed-source do so for
these reasons:

1) they need some proprietary software that doesn't run on Free (e.g.
SAP, PeopleSoft, Siebel, Infor, Baan.....)

2) they need additional functionality (e.g. that shopping cart you
built on LAMP, now needs to talk to MasterCard, so you need PCI and
SOX compliance)

3) they need additional scaling / disaster recovery / high
availability (here we go again: MySQL Cluster does scale horizontally,
but it has its own set of limitations not least of which is it is
memory-based so if you have five years' worth of transactions it may
not fit in memory)

4) the CIO got nervous about running the heart of the business on Free software
Post by Winelfred G. Pasamba
or do those clients who can pay for an oracle license belong to the
market where SAP, Oracle, & others dominate?
Oracle comes in several flavors. Standard Edition and SE1 (which
bundle RAC for free) cost about the same as MS SQL Server and even
less than the cost of Enterprise Support for MySQL.

Enterprise Edition costs a whole lot. But when you consider that a SAP
or PeopleSoft deployment is in the multi-millions of dollars, the
Oracle licensing cost is far from the largest cost.

A certain large local bank recently bought millions (of $) worth of
Oracle for their card system... but they bought more than 2X as much
worth of hardware and who knows how much for the application itself
(the DB and hardware are just infrastructure).
Post by Winelfred G. Pasamba
why is oracle focusing on linux?  does this mean big companies are
starting to not-want-to-pay other platforms?
No it's because Oracle can control the entire stack: virtualization
manager, operating system, database, middle tier, and application.

With other OS you have multiple necks to strangle, so some companies
want to talk to just one supplier, although that's not usually the
case.
Post by Winelfred G. Pasamba
of course we know that the percent of product revenue is going down in
comparison to support revenue, except maybe for microsoft and adobe.*
This is not true. The bulk of Oracle's revenue still comes from license sales.
Post by Winelfred G. Pasamba
is it still true that IBM is still the largest database vendor?*
This has not been true for at least ten years.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Winelfred G. Pasamba
2009-03-26 08:10:04 UTC
Permalink
Post by Orlando Andico
Post by Winelfred G. Pasamba
of course we know that the percent of product revenue is going down in
comparison to support revenue, except maybe for microsoft and adobe.*
This is not true. The bulk of Oracle's revenue still comes from license sales.
i'm looking at professor cusumano's book copyrighted 2004. he claims
that oracle's 2003 income statement says that the company made $9,475M
in revenues and that 35% is from new license fees and 65% is from
services & maintenance. so there has to be an explanation between
your authoritative statement and his interpretation.
Post by Orlando Andico
Post by Winelfred G. Pasamba
is it still true that IBM is still the largest database vendor?*
This has not been true for at least ten years.
my mistake. it is not largest database vendor. IBM is #1 in Software
Magazine's Top Fifty Software Companies (2002) with $47,895M in
software and services revenues. ms is 24,666, eds 21,543, accenture
13,348, and oracle 10,860 (all in millions of dollars).

ibm's total revenues in 2002 was 81,186 including hardware (27,456),
other (4,296), services (36,360) and software (13,074).

anyway it is nice to know that the big guys are backing linux and that
these big guys are somehow making free versions of their great
products which adds to the flavors of databases we can choose from.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-26 09:17:36 UTC
Permalink
winefred, i also stand corrected.

You are right of course in saying that roughly 35% of revenue is new
license sales. Or about $1.5B of the $7B which oracle booked in the
just-concluded quarter.

Here's the thing --- all installed oracle customers must pay an annual
support contract equal to 22% of the original license cost. This
entitles them to free updates and access to fixes/patches for as long
as they maintain the contract.

As you can imagine, this support fee is considerable, it's like buying
a new set of licenses every 5 years -- and paying this support does
NOT entitle you to onsite support. So strictly speaking this is
support & services revenue but it's "by default" -- it comes in
whether or not a new sale is made.

Lots of customers let their support contracts lapse because of the
cost -- they say "i'm paying millions a year but if i want onsite
support i have to pay even more??!!!" but if you're off support and
hit a business-killing bug, you're in trouble.

In any case, Oracle is a very different company today than in 2004.
The large acquisitions -- PeopleSoft, JD Edwards, Siebel Systems,
Hyperion, BEA -- all happened in 2005 onwards. Oracle today is not
just #1 in database, but #1 is middleware, enterprise HR (Peoplesoft)
and CRM (Siebel).
Post by Winelfred G. Pasamba
Post by Orlando Andico
Post by Winelfred G. Pasamba
of course we know that the percent of product revenue is going down in
comparison to support revenue, except maybe for microsoft and adobe.*
This is not true. The bulk of Oracle's revenue still comes from license sales.
i'm looking at professor cusumano's book copyrighted 2004. he claims
that oracle's 2003 income statement says that the company made $9,475M
in revenues and that 35% is from new license fees and 65% is from
services & maintenance. so there has to be an explanation between
your authoritative statement and his interpretation.
Post by Orlando Andico
Post by Winelfred G. Pasamba
is it still true that IBM is still the largest database vendor?*
This has not been true for at least ten years.
my mistake. it is not largest database vendor. IBM is #1 in Software
Magazine's Top Fifty Software Companies (2002) with $47,895M in
software and services revenues. ms is 24,666, eds 21,543, accenture
13,348, and oracle 10,860 (all in millions of dollars).
ibm's total revenues in 2002 was 81,186 including hardware (27,456),
other (4,296), services (36,360) and software (13,074).
anyway it is nice to know that the big guys are backing linux and that
these big guys are somehow making free versions of their great
products which adds to the flavors of databases we can choose from.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
joebert jacaba
2009-03-27 16:31:40 UTC
Permalink
Post by Orlando Andico
Generally people moving from Free software to closed-source do so for
1) they need some proprietary software that doesn't run on Free (e.g.
SAP, PeopleSoft, Siebel, Infor, Baan.....)
why not the owners of these try to make this run on "Free"
Post by Orlando Andico
2) they need additional functionality (e.g. that shopping cart you
built on LAMP, now needs to talk to MasterCard, so you need PCI and
SOX compliance)
open card network?

can these laws be made fair to "Free"?
Post by Orlando Andico
3) they need additional scaling / disaster recovery / high
availability (here we go again: MySQL Cluster does scale horizontally,
but it has its own set of limitations not least of which is it is
memory-based so if you have five years' worth of transactions it may
not fit in memory)
4) the CIO got nervous about running the heart of the business on Free software
i have yet to see a Closed comapny not to use this as a marketing tool.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-28 02:23:20 UTC
Permalink
Post by joebert jacaba
Post by Orlando Andico
Generally people moving from Free software to closed-source do so for
1) they need some proprietary software that doesn't run on Free (e.g.
SAP, PeopleSoft, Siebel, Infor, Baan.....)
why not the owners of these try to make this run on "Free"
Joebert, these enterprise applications are completely non-Free. So the
people who wrote these applications have zero initiative to make them
run on Free.

You think Oracle DB EE cost is prohibitive? then you should look at
the cost of SAP ERP... :-)

Of course there are open-source ERP such as SugarCRM. But a good ERP
solution requires deep knowledge of business rules and transactions.
For say a car manufacturer, that's not knowhow you can pick up off the
street.

The Free software model has worked well for baseline infrastructure
because there's enough skillset out there that the "many eyeballs"
approach works. But as you go up the value chain, more and more domain
knowledge is required and Free starts falling apart.

Let's even look at MySQL Cluster -- the main architect, Mikael
Ronstrom, is a Ph.D. and has devoted the past 15 years of his life to
developing MySQL Cluster. He didn't start out in Free software, he
just had a dream. Good for him Ericsson, MySQL, and now Sun are
willing to pay his salary so he can chase his dream. But the
algorithms in MySQL Cluster are sufficiently complex that there isn't
a large community helping him build it -- it's still mostly Mikael.

Clearly, that's one instance where I think the Free way of software
development has fallen short.
Post by joebert jacaba
Post by Orlando Andico
2) they need additional functionality (e.g. that shopping cart you
built on LAMP, now needs to talk to MasterCard, so you need PCI and
SOX compliance)
open card network?
can these laws be made fair to "Free"?
The open card network is a non-starter. Anyone who's worked in
financial services here knows that a single bank is already a study in
conservatism. Consider that American Express is a distant third place
to Mastercard and Visa. A lot of places here in PH don't even accept
American Express. How a "fourth party" "open card network" would do
better than Amex which has been around for 30+ years, eludes my
comprehension.
Post by joebert jacaba
Post by Orlando Andico
4) the CIO got nervous about running the heart of the business on Free software
i have yet to see a Closed comapny not to use this as a marketing tool.
Guess what... Free software support companies such as Red Hat and
Novell/SuSE also use a variant of this as a marketing tool ("our
billion-dollar company will GUARANTEE that your Free software is up to
snuff!")

The fact that people buy into it means that the proposition has value.
Maybe the systems you're familiar with aren't critical enough (or
produce enough money) that you can't see the value..
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Peter Santiago
2009-03-28 14:56:01 UTC
Permalink
Post by Orlando Andico
Post by joebert jacaba
Post by Orlando Andico
Generally people moving from Free software to closed-source do so for
1) they need some proprietary software that doesn't run on Free (e.g.
SAP, PeopleSoft, Siebel, Infor, Baan.....)
why not the owners of these try to make this run on "Free"
Joebert, these enterprise applications are completely non-Free. So the
people who wrote these applications have zero initiative to make them
run on Free.
You think Oracle DB EE cost is prohibitive? then you should look at
the cost of SAP ERP... :-)
Of course there are open-source ERP such as SugarCRM. But a good ERP
solution requires deep knowledge of business rules and transactions.
For say a car manufacturer, that's not knowhow you can pick up off the
street.
The Free software model has worked well for baseline infrastructure
because there's enough skillset out there that the "many eyeballs"
approach works. But as you go up the value chain, more and more domain
knowledge is required and Free starts falling apart.
Let's even look at MySQL Cluster -- the main architect, Mikael
Ronstrom, is a Ph.D. and has devoted the past 15 years of his life to
developing MySQL Cluster. He didn't start out in Free software, he
just had a dream. Good for him Ericsson, MySQL, and now Sun are
willing to pay his salary so he can chase his dream. But the
algorithms in MySQL Cluster are sufficiently complex that there isn't
a large community helping him build it -- it's still mostly Mikael.
Clearly, that's one instance where I think the Free way of software
development has fallen short.
Post by joebert jacaba
Post by Orlando Andico
2) they need additional functionality (e.g. that shopping cart you
built on LAMP, now needs to talk to MasterCard, so you need PCI and
SOX compliance)
open card network?
can these laws be made fair to "Free"?
The open card network is a non-starter. Anyone who's worked in
financial services here knows that a single bank is already a study in
conservatism. Consider that American Express is a distant third place
to Mastercard and Visa. A lot of places here in PH don't even accept
American Express. How a "fourth party" "open card network" would do
better than Amex which has been around for 30+ years, eludes my
comprehension.
Post by joebert jacaba
Post by Orlando Andico
4) the CIO got nervous about running the heart of the business on Free software
i have yet to see a Closed comapny not to use this as a marketing tool.
Guess what... Free software support companies such as Red Hat and
Novell/SuSE also use a variant of this as a marketing tool ("our
billion-dollar company will GUARANTEE that your Free software is up to
snuff!")
The fact that people buy into it means that the proposition has value.
Maybe the systems you're familiar with aren't critical enough (or
produce enough money) that you can't see the value..
First of all, I'm a lurker here in PLUG. ^_^ This particular thread
have really gone the long way. Bottom line, pragmatism. Use whatever
works for the situation at hand. While saving money is certainly a big
come on for a CIO or CEO, one has to look at the big picture.
Especially during this recession-ridden time, think of what is most cost
effective in the long term. Open source hard asses often forget ROI.
One frequently sees the short term advantages, which unfortunately
doesn't always equate to long-term advantages. Use the right tools for
the job, be it open sourced or closed source. Hopefully we can end this
inane thread before somebody blows a valve.
--
Peter Santiago peters-4/***@public.gmane.org
My website: www.psinergybbs.com
My spamtrap address: r34987y-4/***@public.gmane.org
joebert jacaba
2009-03-30 14:25:01 UTC
Permalink
Post by Orlando Andico
Joebert, these enterprise applications are completely non-Free. So the
people who wrote these applications have zero initiative to make them
run on Free.
they insist on ignoring the open source as an initiative of course.
Post by Orlando Andico
You think Oracle DB EE cost is prohibitive? then you should look at
the cost of SAP ERP...  :-)
which is better then? does cost determine functionality? since SAP is
well known than Oracle in ERP to the enterprise in does that determine
cost?
Post by Orlando Andico
Of course there are open-source ERP such as SugarCRM. But a good ERP
solution requires deep knowledge of business rules and transactions.
For say a car manufacturer, that's not knowhow you can pick up off the
street.
The Free software model has worked well for baseline infrastructure
because there's enough skillset out there that the "many eyeballs"
approach works. But as you go up the value chain, more and more domain
knowledge is required and Free starts falling apart.
it is baseline today. but it is building up and the value chain i
believe will collapse in the future.
Post by Orlando Andico
Let's even look at MySQL Cluster -- the main architect, Mikael
Ronstrom, is a Ph.D. and has devoted the past 15 years of his life to
developing MySQL Cluster. He didn't start out in Free software, he
just had a dream. Good for him Ericsson, MySQL, and now Sun are
willing to pay his salary so he can chase his dream. But the
algorithms in MySQL Cluster are sufficiently complex that there isn't
a large community helping him build it -- it's still mostly Mikael.
Clearly, that's one instance where I think the Free way of software
development has fallen short.
With someone willing to pay your salary or not, nothing should limit
you from achieving your dream. I guess we need more Mikael then.
Post by Orlando Andico
The open card network is a non-starter. Anyone who's worked in
financial services here knows that a single bank is already a study in
conservatism. Consider that American Express is a distant third place
to Mastercard and Visa. A lot of places here in PH don't even accept
American Express. How a "fourth party" "open card network" would do
better than Amex which has been around for 30+ years, eludes my
comprehension.
because that is revolution is all about. making conventional thinking obsolete.
Post by Orlando Andico
Guess what... Free software support companies such as Red Hat and
Novell/SuSE also use a variant of this as a marketing tool ("our
billion-dollar company will GUARANTEE that your Free software is up to
snuff!")
we need more companies like RedHat or Novell/Suse then. :)
Post by Orlando Andico
The fact that people buy into it means that the proposition has value.
Maybe the systems you're familiar with aren't critical enough (or
produce enough money) that you can't see the value..
yes i don't see money as the first bulletpoint to value.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-30 16:34:30 UTC
Permalink
On Mon, Mar 30, 2009 at 10:25 PM, joebert jacaba <joebertj-***@public.gmane.org> wrote:
..
Post by joebert jacaba
Post by Orlando Andico
You think Oracle DB EE cost is prohibitive? then you should look at
the cost of SAP ERP...  :-)
which is better then? does cost determine functionality? since SAP is
well known than Oracle in ERP to the enterprise in does that determine
cost?
Merely pointing out that Oracle is far from the only or worst offender
in the "charge 'em high!" software sweepstakes.

Oracle also does contribute quite a lot to the Linux kernel (around
4th ranked I believe) including BtrFS and AIO.


..
Post by joebert jacaba
Post by Orlando Andico
The Free software model has worked well for baseline infrastructure
because there's enough skillset out there that the "many eyeballs"
approach works. But as you go up the value chain, more and more domain
knowledge is required and Free starts falling apart.
it is baseline today. but it is building up and the value chain i
believe will collapse in the future.
A common fallacy. While it is true that the baseline is continually
going up, the value-add is also going up. There are simply some areas
of expertise that nobody would want to do in their free time.



..
Post by joebert jacaba
With someone willing to pay your salary or not, nothing should limit
you from achieving your dream. I guess we need more Mikael then.
Yes but my dream is to have a nice-enough life, a couple kids, etc.

i don't want to devote 15 years of my life pursuing a utopian
(dystopian?) dream.

Richard Stallman had his MacArthur Prize for him to live off the
interest of so he doesn't need another job.

Finland has a cradle-to-grave social security system and so Linux
could "petiks" all throughout college and grad school while working on
his baby Linux.

I am not denigrating these people's contributions. But they mostly
came from First World countries that could afford to subsidize their
dreams. I don't see that much in the way of OSS contributors locally.

An old friend of mine, Gerard Java, wrote a piece of software that's
literally in every Linux distribution - iptraf. I'm sure you've heard
of it.

Where is he now? in Canada. Because this country couldn't give him a
comfortable-enough livelihood. So what did iptraf get him? I don't
know. Andre Varon (andrelst) is also a friend of his. He's also in
Canada.


..
How a "fourth party" "open card network" would do
Post by joebert jacaba
Post by Orlando Andico
better than Amex which has been around for 30+ years, eludes my
comprehension.
because that is revolution is all about. making conventional thinking obsolete.
Motherhood statement.

As a previous sender mentioned, this thread is getting nowhere. Yes,
you ARE not open-minded, Joebert.


..
Post by joebert jacaba
Post by Orlando Andico
Guess what... Free software support companies such as Red Hat and
Novell/SuSE also use a variant of this as a marketing tool ("our
billion-dollar company will GUARANTEE that your Free software is up to
snuff!")
we need more companies like RedHat or Novell/Suse then. :)
Do you know how much money they charge?

I'm not laughing.


BTW I don't have my own business like Johann does. Tried that, open
source advocacy and solutioning. Didn't fly. We ran out of cash. Bad
business mindedness.

I work for Larry Ellison. While that may in your eyes mark me as a
shill, I don't see it that way. Whether you're working in open or
closed source, you can and do contribute to the community.

And, being on the other side of the fence has showed me how much
talent and effort there is. Closed source companies are not parasites
or "conventional thinking" to be overthrown. I truly believe that most
innovation happens in closed source companies, simply because they
have the funds to invest in new ways of thinking, new technology.

in a way Ballmer was right -- that closed-source companies do most of
the innovation. He was wrong in claiming that Microsoft did most of
the innovation.

If you disagree, please give some examples of software innovation
which truly came from the Free software community.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-30 16:55:28 UTC
Permalink
I'm now getting on my soapbox. This is so OT but what the heck...


After re-reading Joebert's postings, it struck me how uncompromising
and dogmatic his stand is.

Then it struck me... I've read this before!!! and I used to applaud
it!!! Richard Stallman's writings!

One of old RMS's favorite saws is that of the infinite sandwich.
Richard says: if you had an infinite sandwich, why would you sell it?
much better to rid the world of hunger by GIVING away your sandwich,
since it doesn't cost anything to duplicate this sandwich.

According to RMS, once the initial cost of developing the software is
done, additional copies cost nothing to produce! so it's immoral to
profit from these additional copies.

As usual with RMS aphorisms, the sandwich is overly simplistic.

The reason why closed-source companies charge for EVERY copy of their
software, is because they need to money to:

1) develop NEW versions of the software with MORE features

2) pay their shareholders

I don't think anybody will deny that everybody needs a decent
livelihood. Guess what... closed-source employees like myself also
need a decent livelihood. And the shareholders in these companies
deserve a decent return on their investment.

If you put your money in a company, or the bank, would you be happy if
it paid no interest, no return? Hmmmm.

RMS claims that any decent software developer can make a living by
customizing and supporting Free software, so closed-source employees
must be lazy incompetent bastards who can't survive in the Free
ecosystem.

Well guess what... I lived in the Free ecosystem for damn well 90% of
my professional career (9 years). I guess I'm not smart enough because
I had to sell out to Larry.

Nobody will deny that non-Free software costs money. But companies
will only charge what the market is willing to pay. Since the market
is willing to pay Oracle, SAP, IBM, Microsoft, Infor, and God knows
who else, then there is nothing immoral about charging those prices.
And the market is willing to pay because they get something for it.

That "something" may not resonate with Joebert or RMS but it does mean
something to these companies.

Let's not forget that all those great innovations that RMS slavishly
re-implemented (his beloved gcc, grep, etc ad infinitum) all came out
of Bell Labs, which (yes it was a Free software culture) was
bankrolled by the largest, most monopolistic company in history, AT&T.

And let's see... the Unix desktop (GNOME and KDE) have been sapping
developer resources for over 10 years.. that's almost a generation...
and still get bitch-slapped by MacOS X. There really are some things
that a "many eyeballs" Free approach isn't good at addressing.

You don't get something for nothing.

And in my opinion, the price of continued innovation in software, is
having to pay for (at least some) of it.


Anyway back off my soapbox. And back to my evil Closed-Source job that
pays the bills. And incidentally helps a lot of people in this
country, by enabling companies and government agencies to do their
jobs better.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
joebert jacaba
2009-03-30 21:39:24 UTC
Permalink
Post by Orlando Andico
The reason why closed-source companies charge for EVERY copy of their
1) develop NEW versions of the software with MORE features
oracle maybe yes but ms office.
Post by Orlando Andico
2) pay their shareholders
the first sale could have paid them enough.
Post by Orlando Andico
I don't think anybody will deny that everybody needs a decent
livelihood. Guess what... closed-source employees like myself also
need a decent livelihood. And the shareholders in these companies
deserve a decent return on their investment.
If you put your money in a company, or the bank, would you be happy if
it paid no interest, no return? Hmmmm.
of course not. its the reason why you have decided to in the first
place. if your purpose is to donate it then its a different story.
Post by Orlando Andico
Nobody will deny that non-Free software costs money. But companies
will only charge what the market is willing to pay. Since the market
is willing to pay Oracle, SAP, IBM, Microsoft, Infor, and God knows
who else, then there is nothing immoral about charging those prices.
And the market is willing to pay because they get something for it.
lets influence the market to pay less and less then.
Post by Orlando Andico
That "something" may not resonate with Joebert or RMS but it does mean
something to these companies.
Let's not forget that all those great innovations that RMS slavishly
re-implemented (his beloved gcc, grep, etc ad infinitum) all came out
of Bell Labs, which (yes it was a Free software culture) was
bankrolled by the largest, most monopolistic company in history, AT&T.
if oracle is willing to bankroll an a rewrite or show code perhaps in
free terms i dont have a problem with that.
Post by Orlando Andico
And let's see... the Unix desktop (GNOME and KDE) have been sapping
developer resources for over 10 years.. that's almost a generation...
and still get bitch-slapped by MacOS X. There really are some things
that a "many eyeballs" Free approach isn't good at addressing.
im satisfied with GNOME. i can't afford a mac anyway. so bitch-slapped
will not apply to me and to many others i think.
Post by Orlando Andico
You don't get something for nothing.
i get my ubuntu desktop for nothing.
Post by Orlando Andico
And in my opinion, the price of continued innovation in software, is
having to pay for (at least some) of it.
someday innovation will come without a price.
Post by Orlando Andico
Anyway back off my soapbox. And back to my evil Closed-Source job that
pays the bills. And incidentally helps a lot of people in this
country, by enabling companies and government agencies to do their
jobs better.
it is expected to help a lot of people in the country. taxes are spent
paying the license isn't it. if its free then you get an award just
like RMS did. ;)
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Johann Vincent Paul Tagle
2009-03-30 23:09:06 UTC
Permalink
Okay, I propose a cease fire. It is clear for a lot of reasons
(philosophies, advocacies, method of arguing, etc) we're never going to
agree. So let's end this thread already, okay?

Good morning! =)

Johann
Post by joebert jacaba
Post by Orlando Andico
The reason why closed-source companies charge for EVERY copy of their
1) develop NEW versions of the software with MORE features
oracle maybe yes but ms office.
Post by Orlando Andico
2) pay their shareholders
the first sale could have paid them enough.
Post by Orlando Andico
I don't think anybody will deny that everybody needs a decent
livelihood. Guess what... closed-source employees like myself also
need a decent livelihood. And the shareholders in these companies
deserve a decent return on their investment.
If you put your money in a company, or the bank, would you be happy if
it paid no interest, no return? Hmmmm.
of course not. its the reason why you have decided to in the first
place. if your purpose is to donate it then its a different story.
Post by Orlando Andico
Nobody will deny that non-Free software costs money. But companies
will only charge what the market is willing to pay. Since the market
is willing to pay Oracle, SAP, IBM, Microsoft, Infor, and God knows
who else, then there is nothing immoral about charging those prices.
And the market is willing to pay because they get something for it.
lets influence the market to pay less and less then.
Post by Orlando Andico
That "something" may not resonate with Joebert or RMS but it does mean
something to these companies.
Let's not forget that all those great innovations that RMS slavishly
re-implemented (his beloved gcc, grep, etc ad infinitum) all came out
of Bell Labs, which (yes it was a Free software culture) was
bankrolled by the largest, most monopolistic company in history, AT&T.
if oracle is willing to bankroll an a rewrite or show code perhaps in
free terms i dont have a problem with that.
Post by Orlando Andico
And let's see... the Unix desktop (GNOME and KDE) have been sapping
developer resources for over 10 years.. that's almost a generation...
and still get bitch-slapped by MacOS X. There really are some things
that a "many eyeballs" Free approach isn't good at addressing.
im satisfied with GNOME. i can't afford a mac anyway. so bitch-slapped
will not apply to me and to many others i think.
Post by Orlando Andico
You don't get something for nothing.
i get my ubuntu desktop for nothing.
Post by Orlando Andico
And in my opinion, the price of continued innovation in software, is
having to pay for (at least some) of it.
someday innovation will come without a price.
Post by Orlando Andico
Anyway back off my soapbox. And back to my evil Closed-Source job that
pays the bills. And incidentally helps a lot of people in this
country, by enabling companies and government agencies to do their
jobs better.
it is expected to help a lot of people in the country. taxes are spent
paying the license isn't it. if its free then you get an award just
like RMS did. ;)
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
--
Web application and database consulting: www.sagadasolutions.com
Zak B. Elep
2009-03-30 16:58:49 UTC
Permalink
Post by Orlando Andico
An old friend of mine, Gerard Java, wrote a piece of software that's
literally in every Linux distribution - iptraf. I'm sure you've heard
of it.
Where is he now? in Canada. Because this country couldn't give him a
comfortable-enough livelihood. So what did iptraf get him? I don't
know. Andre Varon (andrelst) is also a friend of his. He's also in
Canada.
That takes me back. I used iptraf all the time, when I was maining
Debian/Ubuntu. It was very useful for tracking the rate at real time,
as opposed to other similar (Web) services like speedtest.net. Now
that I main OpenBSD, I use iftop instead, though I miss iptraf's other
features.

I only once met Gerard, at the 2005 LinuxWorld. If I ever meet him
again, I'd buy him a beer or two for his great software.
--
Zak B. Elep || zakame.net
1486 7957 454D E529 E4F1 F75E 5787 B1FD FA53 851D
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-30 17:03:04 UTC
Permalink
Zak:

then I guess the answer is.. iptraf gets you a free beer :-)



On Tue, Mar 31, 2009 at 12:58 AM, Zak B. Elep <zakame-***@public.gmane.org> wrote:
..
That takes me back.  I used iptraf all the time, when I was maining
Debian/Ubuntu.  It was very useful for tracking the rate at real time,
as opposed to other similar (Web) services like speedtest.net.  Now
that I main OpenBSD, I use iftop instead, though I miss iptraf's other
features.
I only once met Gerard, at the 2005 LinuxWorld.  If I ever meet him
again, I'd buy him a beer or two for his great software.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
andrelst
2009-04-20 16:41:17 UTC
Permalink
Small Trivia last time we talked when I was working with an
ISP(franchise of Mozcom) which was eons ago.

* Looking at rewriting iptraf in perl... to be portable.
* He is not subscribed to this list.
* When he discussed to me about iptraf from the start, told him why
when when you can cobble different software to get the same
functionality. Good thing he did not listen to me. :)

regards,
Andre | http://www.varon.ca
Post by Zak B. Elep
Post by Orlando Andico
An old friend of mine, Gerard Java, wrote a piece of software that's
literally in every Linux distribution - iptraf. I'm sure you've heard
of it.
Where is he now? in Canada. Because this country couldn't give him a
comfortable-enough livelihood. So what did iptraf get him? I don't
know. Andre Varon (andrelst) is also a friend of his. He's also in
Canada.
That takes me back. I used iptraf all the time, when I was maining
Debian/Ubuntu. It was very useful for tracking the rate at real time,
as opposed to other similar (Web) services like speedtest.net. Now
that I main OpenBSD, I use iftop instead, though I miss iptraf's other
features.
I only once met Gerard, at the 2005 LinuxWorld. If I ever meet him
again, I'd buy him a beer or two for his great software.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph

joebert jacaba
2009-03-30 21:20:27 UTC
Permalink
Post by Orlando Andico
Merely pointing out that Oracle is far from the only or worst offender
in the "charge 'em high!" software sweepstakes.
at least you admitted that indeed they charge high.
Post by Orlando Andico
Oracle also does contribute quite a lot to the Linux kernel (around
4th ranked I believe) including BtrFS and AIO.
commendable!
Post by Orlando Andico
..
A common fallacy. While it is true that the baseline is continually
going up, the value-add is also going up.
the gap may not really close but it will be closer and closer. the
future will be the judge.
Post by Orlando Andico
There are simply some areas
of expertise that nobody would want to do in their free time.
leaving some areas of expertise that somebody would like to do in
their free time as well.
Post by Orlando Andico
..
Yes but my dream is to have a nice-enough life, a couple kids, etc.
still not nice-enough life?
Post by Orlando Andico
i don't want to devote 15 years of my life pursuing a utopian
(dystopian?) dream.
ok your choice.
Post by Orlando Andico
Richard Stallman had his MacArthur Prize for him to live off the
interest of so he doesn't need another job.
he's deserving anyway.
Post by Orlando Andico
Finland has a cradle-to-grave social security system and so Linux
could "petiks" all throughout college and grad school while working on
his baby Linux.
I am not denigrating these people's contributions. But they mostly
came from First World countries that could afford to subsidize their
dreams. I don't see that much in the way of OSS contributors locally.
it is easy to blame the country eh. wouldn't it be more exciting to do
a Linus while in the Philippines?
Post by Orlando Andico
An old friend of mine, Gerard Java, wrote a piece of software that's
literally in every Linux distribution - iptraf. I'm sure you've heard
of it.
Where is he now? in Canada. Because this country couldn't give him a
comfortable-enough livelihood. So what did iptraf get him? I don't
know. Andre Varon (andrelst) is also a friend of his. He's also in
Canada.
they have their reasons. i don't earn as much as you but i can say im
having a comfortable-enough livelihood. i think its not what will you
get from giving but how many got it and make use of it.
Post by Orlando Andico
..
Motherhood statement.
As a previous sender mentioned, this thread is getting nowhere. Yes,
you ARE not open-minded, Joebert.
you are a participant to this thread dont forget. you are not open-minded, Orly.
Post by Orlando Andico
..
Do you know how much money they charge?
I'm not laughing.
i dont cause im not paying.
Post by Orlando Andico
BTW I don't have my own business like Johann does. Tried that, open
source advocacy and solutioning. Didn't fly. We ran out of cash. Bad
business mindedness.
get rich first then go back to back to advocacy.
Post by Orlando Andico
I work for Larry Ellison. While that may in your eyes mark me as a
shill, I don't see it that way.
i like Larry Ellison until he got married.
Post by Orlando Andico
Whether you're working in open or
closed source, you can and do contribute to the community.
i dont doubt your contributions sir.
Post by Orlando Andico
And, being on the other side of the fence has showed me how much
talent and effort there is. Closed source companies are not parasites
or "conventional thinking" to be overthrown. I truly believe that most
innovation happens in closed source companies, simply because they
have the funds to invest in new ways of thinking, new technology.
in a way Ballmer was right -- that closed-source companies do most of
the innovation. He was wrong in claiming that Microsoft did most of
the innovation.
If you disagree, please give some examples of software innovation
which truly came from the Free software community.
i agree. software patents?
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Ariz Jacinto
2009-03-28 00:22:40 UTC
Permalink
Based on your example below, I understand why companies would want to
use proprietary software because the available turn key solution[1]
appeals to them and _not_ because it is a requirement for PCI-DSS[2]
compliance. Maybe it's not what you really meant but the statement
could easily be taken out of context.

So just to be clear with the uninitiated, PCI security standard does
not discriminate against the use of open-source software. Meaning, a
LAMP-based shopping cart could still connect to various payment
gateways either directly or via third-party and can still be
PCI-compliant.

[1] https://www.mastercardpaymentgateway.com/mpgpublic/faq.do#q21
[2] https://pcisecuritystandards.org/security_standards/download.html?id=pci_dss_v1-2.pdf
Post by Orlando Andico
...
Generally people moving from Free software to closed-source do so for
...
2) they need additional functionality (e.g. that shopping cart you
built on LAMP, now needs to talk to MasterCard, so you need PCI and
SOX compliance)
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-28 02:11:58 UTC
Permalink
Agreed.. I did mention that it's possible to be PCI compliant without
using a closed solution. But it would be much harder (for example the
data user / DBA user role separation requirement seems -- at first
glance, to me -- not possible using MySQL). Even if it were possible,
you'd have a lot more integration to do.

Some of Joebert's ideas (such as an open card network) seem completely
impractical.

Anyway Johann said it very elegantly. On one side you have people like
Ballmer. On the other side... Richard Stallman and Joebert :-)

I believe everyone acknowledges Richard Stallman's contributions,
although I believe everyone also believes he's an annoying guy
sometimes (GNU/Linux, give it a rest..) So there will always be people
who would insist on a certain approach on religious grounds.

I'd like to believe I'm somewhere in between these two extremes
(although I used to be very religious about Free... I guess as you get
older and work with more people...)
Post by Ariz Jacinto
Based on your example below, I understand why companies would want to
use proprietary software because the available turn key solution[1]
appeals to them and _not_ because it is a requirement for PCI-DSS[2]
compliance. Maybe it's not what you really meant but the statement
could easily be taken out of context.
So just to be clear with the uninitiated, PCI security standard does
not discriminate against the use of open-source software. Meaning, a
LAMP-based shopping cart could still connect to various payment
gateways either directly or via third-party and can still be
PCI-compliant.
[1] https://www.mastercardpaymentgateway.com/mpgpublic/faq.do#q21
[2] https://pcisecuritystandards.org/security_standards/download.html?id=pci_dss_v1-2.pdf
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
joebert jacaba
2009-03-27 16:14:56 UTC
Permalink
On Wed, Mar 25, 2009 at 9:56 AM, Johann Vincent Paul Tagle
Post by Johann Vincent Paul Tagle
I don't think they are thinking short term but thinking what makes sense
with regard to their business.  They need such a feature now, they can pay
for it, so they go with what's available and can be reliably supported now.
clearly short term since you used the word "now."
Post by Johann Vincent Paul Tagle
One of your objectives in your biking must be to get healthy, and there's
really no shortcut to it but dedication and practice.  However, in business,
there are many ways to accomplish something, and what you choose should be
what makes sense to you.  If you are making say 25M per year on a product
using an Oracle database, the initial 2M investment and the succeeding 0.5M
yearly support cost is nothing.
Anything as long as you earn?
Post by Johann Vincent Paul Tagle
Even if in the long run the open source
alternative is a lot cheaper, they won't be willing to make the change,
because what they have simply works for them.
You used "long run". Are you thinking the other one is short term then?
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Johann Vincent Paul Tagle
2009-03-27 23:58:22 UTC
Permalink
Post by joebert jacaba
On Wed, Mar 25, 2009 at 9:56 AM, Johann Vincent Paul Tagle
Post by Johann Vincent Paul Tagle
I don't think they are thinking short term but thinking what makes sense
with regard to their business. They need such a feature now, they can
pay
Post by Johann Vincent Paul Tagle
for it, so they go with what's available and can be reliably supported
now.
clearly short term since you used the word "now."
Using the word "now" doesn't mean short term. Just what makes sense for
business. It won't make sense for me to wait 6-12 months to start earning
say 2M/month when I can do so in 2 months if I invest 2M now. And it
doesn't make sense to change what works as long as its cost is within
acceptable limits.
Post by joebert jacaba
Post by Johann Vincent Paul Tagle
One of your objectives in your biking must be to get healthy, and there's
really no shortcut to it but dedication and practice. However, in
business,
Post by Johann Vincent Paul Tagle
there are many ways to accomplish something, and what you choose should
be
Post by Johann Vincent Paul Tagle
what makes sense to you. If you are making say 25M per year on a product
using an Oracle database, the initial 2M investment and the succeeding
0.5M
Post by Johann Vincent Paul Tagle
yearly support cost is nothing.
Anything as long as you earn?
Yes. As long as it makes business sense. Now you may say it makes business
sense to bring the cost down even more. True, but to some, a certain level
of cost is acceptable as long as there are benefits (e.g. a big company to
turn to).
Post by joebert jacaba
Post by Johann Vincent Paul Tagle
Even if in the long run the open source
alternative is a lot cheaper, they won't be willing to make the change,
because what they have simply works for them.
You used "long run". Are you thinking the other one is short term then?
No. Just comparing costs.

Joebert, judging from your posts you are an open source champion. You will
do anything to promote use of free and open source solutions. I respect
that deeply. However, as you said in a previous post "Use what works for
you, your client, your brain, your pocket, your advocacy." Right tool for
right job. So let's leave it there. I watched this thread morph from a
technical discussion of MySQL and Oracle into a talk of business
philosophies, etc.

Suggestion ko lang, open your mind up a little - you seem to be promoting
open source in a similar way how MS did for closed source before - e.g. "the
alternative is just wrong/evil". You'll reach more people if you don't
readily dismiss certain features as "bling" - just imagine pitching a
solution to somebody then dismissing the alternative as bling, when the
person you're pitching to is somebody who recognizes the benefits of both
open and closed source solutions. He will likely not want to listen to you
anymore. People like these want to talk to those who either share the same
views, or those who just focus on the benefits of the solution they are
proposing and let the client decide for themselves.

Good luck in your advocacy.

Johann
--
Web application and database consulting: www.sagadasolutions.com
joebert jacaba
2009-03-30 14:55:19 UTC
Permalink
On Sat, Mar 28, 2009 at 7:58 AM, Johann Vincent Paul Tagle
Using the word "now" doesn't mean short term.  Just what makes sense for
business.  It won't make sense for me to wait 6-12 months to start earning
say 2M/month when I can do so in 2 months if I invest 2M now.  And it
doesn't make sense to change what works as long as its cost is within
acceptable limits.
as long as the price tag is M then "now" can be redefined e.
Yes.  As long as it makes business sense.  Now you may say it makes business
sense to bring the cost down even more.  True, but to some, a certain level
of cost is acceptable as long as there are benefits (e.g. a big company to
turn to).
the benefit you cited is not to me. kawawa naman philippines we don't
have this big companies. how can we compete then if all of these
talented individuals locally is just contented to be salaried by these
big companies.
No.  Just comparing costs.
if you can redefine "now" as not short then you can do that with
"long" too. money goes a long way.
Joebert, judging from your posts you are an open source champion.  You will
do anything to promote use of free and open source solutions.  I respect
that deeply.  However, as you said in a previous post "Use what works for
you, your client, your brain, your pocket, your advocacy."  Right tool for
right job.  So let's leave it there.  I watched this thread morph from a
technical discussion of MySQL and Oracle into a talk of business
philosophies, etc.
people will do anything as well for money. :)
Suggestion ko lang, open your mind up a little
ako pa ngayon ang di open. and people have been defending closed here.
hehe. im not always right and i hope you also accept that fact.
- you seem to be promoting
open source in a similar way how MS did for closed source before - e.g. "the
alternative is just wrong/evil".
didn't they got rich using that? everything changes e?
You'll reach more people if you don't
readily dismiss certain features as "bling" - just imagine pitching a
solution to somebody then dismissing the alternative as bling, when the
person you're pitching to is somebody who recognizes the benefits of both
open and closed source solutions.  He will likely not want to listen to you
anymore.
youre still listening pa naman. its all right then.
People like these want to talk to those who either share the same
views, or those who just focus on the benefits of the solution they are
proposing and let the client decide for themselves.
if we share the same views will just praise each other. not good.
Good luck in your advocacy.
Good luck in your business. Hope you (and Orly) get very rich soon and
side with me. Just like Shuttleworth.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Kelsey Hartigan Go
2009-03-19 10:46:05 UTC
Permalink
There is a free version of Oracle for development...but severely
restricted...similar to microsoft sqlexpress.and Db2express (but db2 and
sqlexpress you can use for production, oracle, you can't)


On Thu, Mar 19, 2009 at 12:05 PM, Winelfred G. Pasamba <
Post by Winelfred G. Pasamba
Is there are possible market state where Oracle will zero the
license-to-use price and only charge for support?
Post by Orlando Andico
Sun actually positions MySQL as an Oracle replacement in some cases,
and you ought to check out their support contract pricing.. the
comparison is more relevant than you guys think.
Post by joebert jacaba
Post by Jerome Gotangco
I think its unfair to compare Oracle 10 or 11g to Postgres. There are
a lot of verticals that Oracle cover that Postgres probably doesn't
because of market demand. A more appropriate comparison would probably
be OracleXE to PostgreSQL.
And is it fair to compare it to MySQL?
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
--
Kelsey Hartigan Go
Registered Linux user #5998
Pablo Manalastas
2009-03-19 11:26:27 UTC
Permalink
I went through the earlier posts and tried the usb_modeswitch
solution on my MSI Wind running Ubuntu Intrepid. Didn't work!
Anyone got this to work on a system like mine? Thanks.

Pablo Manalastas

_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Erwin Olario
2009-03-19 11:30:27 UTC
Permalink
Here <http://ubuntuforums.org/archive/index.php/t-1017630.html> is someone
using the same modem on Intrepid.
Post by Pablo Manalastas
I went through the earlier posts and tried the usb_modeswitch
solution on my MSI Wind running Ubuntu Intrepid. Didn't work!
Anyone got this to work on a system like mine? Thanks.
Pablo Manalastas
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Michael Tinsay
2009-03-19 13:56:26 UTC
Permalink
Doc,

You'll need the 2.6.28 kernel available from the Jaunty repos. The ZTE 3G USB modem drivers apparently made it to the kernel only starting from 2.6.27.


--- mike t.




----- Original Message ----
Sent: Thursday, March 19, 2009 19:26:27
Subject: [plug] Smart Bro prepaid (ZTE MF627) USB modem in Linux
I went through the earlier posts and tried the usb_modeswitch
solution on my MSI Wind running Ubuntu Intrepid. Didn't work!
Anyone got this to work on a system like mine? Thanks.
Pablo Manalastas
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Pablo Manalastas
2009-03-19 21:54:24 UTC
Permalink
My Smart Bro ZTE MF627 modem now works properly for my
MSI Wind running Ubuntu Intrepid. During my first unsuccessful
attempt at installing usb_modeswitch Debian package, we are
supposed to comment out the first entry in /etc/usb_modeswitch.conf
and uncomment the entry for ZTE MF627 (or 628+). I forgot to
comment out the first entry.

Now that I followed the instruction correctly, I have the
ZTE USB stick working under NetworkManager. I just miss the
details of connection information that is available under
Windows that is not supplied by NetworkManager, like how much
time I have already consumed online.

What is the proper way to disconnect so that I do not get charged
more than I actually used?

Thanks to all who replied to my post.

Pablo Manalastas
***
Subject: [plug] Smart Bro prepaid (ZTE MF627) USB modem in Linux
Date: Thursday, March 19, 2009, 7:26 PM
I went through the earlier posts and tried the
usb_modeswitch
solution on my MSI Wind running Ubuntu Intrepid.
Didn't work!
Anyone got this to work on a system like mine? Thanks.
Pablo Manalastas
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Pablo Manalastas
2009-03-20 02:19:24 UTC
Permalink
I am at McDonald's Katipunan, in front of Ateneo Gate 2.
Free WiFi service is at best choppy - you're sometimes connected
and oftentimes disconnected. I do not know how McDo does it,
but this connect-disconnect phenomenon never happens with the
Linksys WiFi router that I have at home. I get 384Kbps at best
at home, but that is a reliable 384Kbps.

So I stick in my Smart Bro USB modem stick into my MSI Wind,
and start downloading OpenOffice 3.0.1 for Windows. I finished
downloading 142M in 14m 28s which is an amazing 168 KB/s, at the
cost of PHP10.00/30 minutes. I get 35 - 42KB/s at home and I pay
PHP999.00 a month. Of course you do not always get this fast
connection (HSDPA) everywhere in Metro Manila. Katipunan just
happens to have HSDPA signal.

I am totally happy with my Smart Bro prepaid USB modem. It is
there when I need Internet connection, and it is reasonably
fast when I am at the correct HSDPA-covered locations.

Galing!

Pablo Manalastas
***
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Holden Hao
2009-03-20 05:09:49 UTC
Permalink
On Fri, Mar 20, 2009 at 10:19 AM, Pablo Manalastas
Post by Pablo Manalastas
I am at McDonald's Katipunan, in front of Ateneo Gate 2.
Free WiFi service is at best choppy - you're sometimes connected
and oftentimes disconnected. I do not know how McDo does it,
but this connect-disconnect phenomenon never happens with the
Most likely it is due to interference. Perhaps there are plenty of users
that are logged in and they are all vying for attention from the poor access
point. So if you have a weak signal then you will be the first to get cut
off.


Holden
Winelfred G. Pasamba
2009-03-19 11:31:51 UTC
Permalink
one of the very nice things about Oracle is RAC.
are there other databases that have similar functionality?

from a simplistic view one wonders if RAC is using the same ideas used
for cache coherency on SMP systems with separate caches on each cpu.

On Thu, Mar 19, 2009 at 6:46 PM, Kelsey Hartigan Go
Post by Kelsey Hartigan Go
There is a free version of Oracle for development...but severely
restricted...similar to microsoft sqlexpress.
and Db2express (but db2 and sqlexpress you can use for production, oracle,
you can't)
On Thu, Mar 19, 2009 at 12:05 PM, Winelfred G. Pasamba
Post by Winelfred G. Pasamba
Is there are possible market state where Oracle will zero the
license-to-use price and only charge for support?
Post by Orlando Andico
Sun actually positions MySQL as an Oracle replacement in some cases,
and you ought to check out their support contract pricing.. the
comparison is more relevant than you guys think.
Post by joebert jacaba
Post by Jerome Gotangco
I think its unfair to compare Oracle 10 or 11g to Postgres. There are
a lot of verticals that Oracle cover that Postgres probably doesn't
because of market demand. A more appropriate comparison would probably
be OracleXE to PostgreSQL.
And is it fair to compare it to MySQL?
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
--
Kelsey Hartigan Go
Registered Linux user #5998
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
--
Seek ye first the kingdom of God and all these things shall be added unto you.

Winelfred G. Pasamba
Adventist University of the Philippines Online Information Systems
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
thad
2009-03-20 03:27:21 UTC
Permalink
Post by Winelfred G. Pasamba
one of the very nice things about Oracle is RAC.
are there other databases that have similar functionality?
from a simplistic view one wonders if RAC is using the same ideas used
for cache coherency on SMP systems with separate caches on each cpu.
Good if you can find open source that scale like RAC. I work for a
company that could afford to spend $2million for 2 P6-P595 on DB2 +
HACMP for high availability but having serious thought for spending
several more millions for Oracle RAC that will fit our requirement. Of
course we UNIX SAs and some Oracle DBAs want RAC. Its not elegant for
the high availability cluster on the OS level to handle a database
fail over scenario, RAC can handle it better. For automatic metering
service and smart grid, Oracle RAC + NMS is the way to go.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
fooler mail
2009-03-20 05:53:37 UTC
Permalink
Post by thad
Post by Winelfred G. Pasamba
one of the very nice things about Oracle is RAC.
Good if you can find open source that scale like RAC.
you have to understand between "shared-everything" and "shared-nothing
" architectures in a distributed system when you are talking about
scalability guys...

oracle rac is a shared-everything architecture while mysql cluster is
a shared-nothing architecture..

the problem with rac is that it shared its backend storage... thus the
bottleneck and scalability problem comes when more nodes added to a
RAC especially when the nodes are doing a write operations...
furthermore.. oracle 11g can support now upto 100 nodes... in real
implementations.. it shows negative scalability because of the nodes
coordinating with each other to communicate who owns this resource
(eg. lock ownership).

with that scalability problem of shared-everything architecture...
here comes the shared-nothing architecture..

shared-nothing architecture where each node is independent and
self-sufficient and there is no single point of contention across the
system... one good example and demonstrated how effective
shared-nothing architecture and scale almost *infinitely* by adding an
inexpensive node since there is no single bottleneck to slow down the
system is google database (using a flat file spreads across the
system)... amazon is doing this also... with a massive read and write
transactions.. shared-nothing is the logical choice here...

i have experienced with oracle rac (10gR2) and experienced how it was
bog down the whole RAC system too.. tsk tsk tsk

there are pros and cons to everything... not siding with joebert as i
know where is coming from... he just simply understand his database
needs and just a practical person that he simply knows what database
works for him... for him.. why buy an apple if i need an orange?

in short.. know your workload first and choose what database is right for you...

fooler.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-20 10:48:15 UTC
Permalink
On Fri, Mar 20, 2009 at 1:53 PM, fooler mail <fooler.mail-***@public.gmane.org> wrote:
..
Post by fooler mail
the problem with rac is that it shared its backend storage... thus the
bottleneck and scalability problem comes when more nodes added to a
RAC especially when the nodes are doing a write operations...
furthermore..  oracle 11g can support now upto 100 nodes... in real
implementations.. it shows negative scalability because of the nodes
coordinating with each other to communicate who owns this resource
(eg. lock ownership).
..
Post by fooler mail
i have experienced with oracle rac (10gR2) and experienced how it was
bog down the whole RAC system too.. tsk tsk tsk
Oracle is not exactly shared-everything. While the storage is indeed a
single point of bottleneck, the existence of Cache Fusion means that
you will not bottleneck the disk right away.

As for lock contention... that's why in a correctly-configured RAC
environment you MUST use the Partitioning option as well (that's an
additional $10K+ / CPU, ka-ching!) and partition your tables properly
so that you won't have lock contention.

To simply say "in real implementatons RAC shows negative scalability"
is a very common misconception and indeed, people see this in practice
but it's because they don't know how to tune RAC.

Shared-nothing is very challenging simply because you don't have a
single system view. Most people don't have to have to aggregate their
data from all over the place just to do a single query.

In any case, in Oracle 11g the new Exadata storage cell concept allows
linear scalability to very many nodes and removes the shared-storage
requirement that RAC has (traditionally) had.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
fooler mail
2009-03-20 13:21:35 UTC
Permalink
Post by Orlando Andico
Oracle is not exactly shared-everything. While the storage is indeed a
single point of bottleneck, the existence of Cache Fusion means that
you will not bottleneck the disk right away.
oracle rac belongs to shared-everything architecture... the cache
fusion is a global cache (distributed caching) similar to disk cache
in a single node to delay writes... still a bottleneck when you are
running out of memory as it needs to write to give room for incoming
data to write when write load increases and you are having a
scalability problem there...
Post by Orlando Andico
As for lock contention... that's why in a correctly-configured RAC
environment you MUST use the Partitioning option as well (that's an
additional $10K+ / CPU, ka-ching!) and partition your tables properly
so that you won't have lock contention.
no matter how you spend more money to improve lock contention.. there
is always a lock contention in a shared-everything architecture simply
because two or more nodes can still access the same block on that
partition..
Post by Orlando Andico
To simply say "in real implementatons RAC shows negative scalability"
is a very common misconception and indeed, people see this in practice
but it's because they don't know how to tune RAC.
naahhh... the internal communication of RAC in a separate channel
simply shows a negative scalability when the nodes increases due to
the fact that each node must coordinate with each other regarding to
lock ownership and all nodes updates the same control file as well as
data-block ownership transfer which creates a big bottleneck in the
performance..
Post by Orlando Andico
Shared-nothing is very challenging simply because you don't have a
single system view.
you do...
Post by Orlando Andico
Most people don't have to have to aggregate their
data from all over the place just to do a single query.
let the index do its job (if a query has a condition) and fetch those
data where it resides to a particular node.. thus less load to every
disk storage on every node and much faster to get the data if you
compare that to a shared storage... that is why google is too damn
fast to return a search query because of a shared-nothing
architecture...
Post by Orlando Andico
In any case, in Oracle 11g the new Exadata storage cell concept allows
linear scalability to very many nodes and removes the shared-storage
requirement that RAC has (traditionally) had.
exadata storage cell concept scales simply because they move their
traditional shared-everything storage like rac to shared-nothing like
exadata storage.. your exadata storage is a shared-nothing
architecture... now as you can see oracle starting to understand and
reliaze how shared-nothing architecture scales...

fooler.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-20 14:27:17 UTC
Permalink
Post by fooler mail
Post by Orlando Andico
Oracle is not exactly shared-everything. While the storage is indeed a
single point of bottleneck, the existence of Cache Fusion means that
you will not bottleneck the disk right away.
oracle rac belongs to shared-everything architecture... the cache
fusion is a global cache (distributed caching) similar to disk cache
in a single node to delay writes... still a bottleneck when you are
running out of memory as it needs to write to give room for incoming
data to write when write load increases and you are having a
scalability problem there...
Yes, I was pointing out that the global cache allows you to delay
writes, which results in better scalability to the disk. The disk
still is a "single point of knowledge" and therefore is, yes, a
shared-all architecture.
Post by fooler mail
Post by Orlando Andico
As for lock contention... that's why in a correctly-configured RAC
environment you MUST use the Partitioning option as well (that's an
additional $10K+ / CPU, ka-ching!) and partition your tables properly
so that you won't have lock contention.
no matter how you spend more money to improve lock contention.. there
is always a lock contention in a shared-everything architecture simply
because two or more nodes can still access the same block on that
partition..
Post by Orlando Andico
To simply say "in real implementatons RAC shows negative scalability"
is a very common misconception and indeed, people see this in practice
but it's because they don't know how to tune RAC.
naahhh... the internal communication of RAC in a separate channel
simply shows a negative scalability when the nodes increases due to
the fact that each node must coordinate with each other regarding to
lock ownership and all nodes updates the same control file as well as
data-block ownership transfer which creates a big bottleneck in the
performance..
Of course taken to the extreme, any shared architecture will show
negative scalability at some point. I was talking about N where N is a
small integer. Going from 1 node to 2, or 2 nodes to 4, will not show
negative scalability in most cases, and Oracle isn't generally used
for large N, regardless of what the product literature claims.
Post by fooler mail
Post by Orlando Andico
Shared-nothing is very challenging simply because you don't have a
single system view.
you do...
Post by Orlando Andico
Most people don't have to have to aggregate their
data from all over the place just to do a single query.
let the index do its job (if a query has a condition) and fetch those
data where it resides to a particular node.. thus less load to every
disk storage on every node and much faster to get the data if you
compare that to a shared storage... that is why google is too damn
fast to return a search query because of a shared-nothing
architecture...
I'm not aware of any off-the-shelf shared-nothing architectures which
allow you to scatter the data over multiple nodes. Except Exadata, as
you point out.

If we count "roll your own," such as Google does, that's not something
readily replicatable by most end-users.
Post by fooler mail
Post by Orlando Andico
In any case, in Oracle 11g the new Exadata storage cell concept allows
linear scalability to very many nodes and removes the shared-storage
requirement that RAC has (traditionally) had.
exadata storage cell concept scales simply because they move their
traditional shared-everything storage like rac to shared-nothing like
exadata storage.. your exadata storage is a shared-nothing
architecture... now as you can see oracle starting to understand and
reliaze how shared-nothing architecture scales...
It is not shared-nothing, because it still is fronted by RAC. The
storage nodes simply abstract the concept of storage to "smart
storage." Yes in a way it is "Federated" which is a bad word in most
places (most "federated" architectures tend to be fragile)... but in
the case of Exadata the federation and storage management is all
internal to the system so you don't have to be a rocket scientist to
build a federated database, it's pure plug-and-play (add storage nodes
on the fly for more scaling).

Like I mentioned above, there's no denying you can build a fast
shared-nothing architecture like Google has done. But is this
something an ordinary user can do? in the case of Exadata, you can (if
you have enough money). It's meaningless to compare this functionality
to MySQL when you'd basically have to build it from zero on one side,
and buy a box on the other side.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
fooler mail
2009-03-20 16:36:37 UTC
Permalink
Post by Orlando Andico
Post by fooler mail
oracle rac belongs to shared-everything architecture... the cache
fusion is a global cache (distributed caching) similar to disk cache
in a single node to delay writes... still a bottleneck when you are
running out of memory as it needs to write to give room for incoming
data to write when write load increases and you are having a
scalability problem there...
Yes, I was pointing out that the global cache allows you to delay
writes, which results in better scalability to the disk. The disk
still is a "single point of knowledge" and therefore is, yes, a
shared-all architecture.
in a shared-nothing architecture... all nodes has its own disk cache
for delay writes but the writes are not pointed to a centralized
storage.. thus a higher throughput and scalability...
Post by Orlando Andico
Post by fooler mail
naahhh... the internal communication of RAC in a separate channel
simply shows a negative scalability when the nodes increases due to
the fact that each node must coordinate with each other regarding to
lock ownership and all nodes updates the same control file as well as
data-block ownership transfer which creates a big bottleneck in the
performance..
Of course taken to the extreme, any shared architecture will show
negative scalability at some point. I was talking about N where N is a
small integer. Going from 1 node to 2, or 2 nodes to 4, will not show
negative scalability in most cases, and Oracle isn't generally used
for large N, regardless of what the product literature claims.
it was claimed for enterprise environment and enterprise expected a
high performance database especially on scaling as they are paying big
bucks for this enterprise database... there is no problem on the
frontend but on the backend.... oracle knows this thus their solution
is exadata storage...
Post by Orlando Andico
Post by fooler mail
let the index do its job (if a query has a condition) and fetch those
data where it resides to a particular node.. thus less load to every
disk storage on every node and much faster to get the data if you
compare that to a shared storage... that is why google is too damn
fast to return a search query because of a shared-nothing
architecture...
I'm not aware of any off-the-shelf shared-nothing architectures which
allow you to scatter the data over multiple nodes. Except Exadata, as
you point out.
mysql cluster is one of it... greenplum... teradata... netezza.. etc..
that is how shared-nothing architecture really does...
Post by Orlando Andico
If we count "roll your own," such as Google does, that's not something
readily replicatable by most end-users.
let us take a look how google does... their algorithm is pretty clean
and simple... it just a one big flat file... the raw data is partition
and spread to all nodes.. another set of server farm where it holds
the one big index file that spread to all nodes on that cluster...
query begins by consulting the index file and the index file pointing
the actual data to a particular node.. the algorithm here is actually
the same with a single node database where the index file pointing the
actual data to a particular offset of a file..
Post by Orlando Andico
Post by fooler mail
exadata storage cell concept scales simply because they move their
traditional shared-everything storage like rac to shared-nothing like
exadata storage.. your exadata storage is a shared-nothing
architecture... now as you can see oracle starting to understand and
reliaze how shared-nothing architecture scales...
It is not shared-nothing, because it still is fronted by RAC.
ill cut and paste the exadata white paper here...

The Exadata products address the three key dimensions of database I/O
that can hamper data warehouse performance.
• Exadata is based on a massively parallel architecture which provides
more pipes to deliver more data faster between the database servers
and the storage servers.
• Exadata is built using wider pipes that provide extremely high
bandwidth between the database servers and the storage servers.
• Exadata is database aware and can ship just the data required to
satisfy SQL requests resulting is less data being sent between the
database servers and the storage servers.

take note that exadata is based on a massively parallel architecture
(MPP)... MPP is actually a shared-nothing architecture..
Post by Orlando Andico
The
storage nodes simply abstract the concept of storage to "smart
storage." Yes in a way it is "Federated" which is a bad word in most
places (most "federated" architectures tend to be fragile)... but in
the case of Exadata the federation and storage management is all
internal to the system so you don't have to be a rocket scientist to
build a federated database, it's pure plug-and-play (add storage nodes
on the fly for more scaling).
but you dont realize the one who built exadata is a rocket
scientist... the same thing you dont need to build a database program
from scratch as another rocket scientist already built it for you..
you just use it...
Post by Orlando Andico
Like I mentioned above, there's no denying you can build a fast
shared-nothing architecture like Google has done. But is this
something an ordinary user can do? in the case of Exadata, you can (if
you have enough money). It's meaningless to compare this functionality
to MySQL when you'd basically have to build it from zero on one side,
and buy a box on the other side.
i dont see any point from you here... any company or a person can
build a product to sell or for free to an ordinary user.. they just
simply create a cd installation disk to automate things... with
exadata it just a finish product...

fooler.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-20 18:16:15 UTC
Permalink
This thread has gone in two directions at once: first, a generic
"shared-nothing" versus "shared everything" debate, and second,
comparing MySQL Cluster to other things.

Since this was originally a "Oracle to MySQL migration" thread I'll
Post by fooler mail
if you said about rac that people dont know how to tune or configure
rac.. they same true also to people who dont know how to configure
mysql cluster...
Very true. Expertise doesn't come free with any technology.
Post by fooler mail
alcatel uses mysql cluster for their home location register (HLR)..
HLR is a central database that contains details of each mobile phone
subscriber that is authorized to use the gsm network.. vodafone in UK
uses mysql cluster too..
I would expect that MySQL Cluster is not a major player in the
embedded telecom space. There are a lot of incumbents in there:
TimesTen, SolidDB, Altibase to name a few.

Ericsson, T-Mobile, China Mobile, Amdocs, Salesforce.com etc. embed
TimesTen. Nokia-Siemens embeds SolidDB. SK Telecom and Samsung embed
Altibase.

NDB Cluster code base came from Ericsson... but they no longer use it
(I guess they decided making databases was not their core business and
someone else could do it better).

Incidentally, if Sun acquisition by IBM pushes through, I don't know
what will be the fate of MySQL Cluster, as IBM owns an existing (and
better-proven in the field) IMDB in the form of SolidDB, not to
mention DB2.
Post by fooler mail
the mysql cluster replication provides highly availability in case one
node fails the other node will take care of that data... usually
replication took place when an update on the master server... it
propagates to a slave server and not to the rest of the nodes...
This is fine, in fact part of the requirement for in-memory store is
so that acknowledges from the slave happen "fast enough" to maintain
throughput. The MySQL Cluster manual states (and I quote) that

"In MySQL Cluster, all data nodes are kept in synchrony, and a
transaction committed by any one data node is committed for all data
nodes. In the event of a data node failure, all remaining data nodes
remain in a consistent state. " (23.10.3)

That seems like "write to one, write to all" although there is an
option for setting the number of replicas. It would seem that due to
the synchronous commit, write performance can't be THAT high.

To be honest, the last time I checked MySQL Cluster was 3 years ago
and I was not impressed. The feature set seems to be improving nicely.
I think it's time I revisit MySQL Cluster. (for which I have fooler to
thank...)
Post by fooler mail
in a shared-nothing architecture... all nodes has its own disk cache
for delay writes but the writes are not pointed to a centralized
storage.. thus a higher throughput and scalability...
This is fine if the workload can be cleanly partitioned. However many
transactional workloads cannot be sliced-and-diced so easily.

Having shared-nothing and individual buffer caches means you cannot
guarantee transactional integrity across all members. For things like
Google (which is bulk write from robot + many many reads) basically
any read-mostly workload, this will work fine.
Post by fooler mail
Post by Orlando Andico
Of course taken to the extreme, any shared architecture will show
negative scalability at some point. I was talking about N where N is a
small integer. Going from 1 node to 2, or 2 nodes to 4, will not show
negative scalability in most cases, and Oracle isn't generally used
for large N, regardless of what the product literature claims.
it was claimed for enterprise environment and enterprise expected a
high performance database especially on scaling as they are paying big
bucks for this enterprise database... there is no problem on the
frontend but on the backend.... oracle knows this thus their solution
is exadata storage...
As N gets large, RAC scalability for writes gets to the point of
diminishing returns. Not a limitation unique to RAC, even SMP boxes
suffer from this.

Shared-nothing is a known scalable solution, almost all the Top500
supercomputers are MPP. But shared-nothing has been traditionally very
hard use get working properly for traditional workloads.
Post by fooler mail
Post by Orlando Andico
I'm not aware of any off-the-shelf shared-nothing architectures which
allow you to scatter the data over multiple nodes. Except Exadata, as
you point out.
mysql cluster is one of it... greenplum... teradata... netezza.. etc..
that is how shared-nothing architecture really does...
Greenplum and Netezza (and DATAllegro) have tiny market share. The
much-derided RAC has over 40% of the worldwide data warehousing
market. So if you went for these products, you'd have a much smaller
community to turn to for help. And these products (well even Exadata)
aren't designed for OLTP workloads which I believe are most
interesting to PLUG readers.
Post by fooler mail
If we count "roll your own," such as Google does, that's not something
readily replicatable by most end-users.
let us take a look how google does... their algorithm is pretty clean
and simple... it just a one big flat file... the raw data is partition
and spread to all nodes.. another set of server farm where it holds
the one big index file that spread to all nodes on that cluster...
query begins by consulting the index file and the index file pointing
the actual data to a particular node.. the algorithm here is actually
the same with a single node database where the index file pointing the
actual data to a particular offset of a file..
I don't see Google wannabes popping up right and left if Google's
implementation was that clean and simple... Google's technology is one
of their competitive advantages and I don't see anyone seriously
threatening them at what they do.

Besides it's built for a query workload where it's permissible to lose
data. Which is an important use case but not one that most of us can
use.
Post by fooler mail
Post by Orlando Andico
exadata storage cell concept scales simply because they move their
traditional shared-everything storage like rac to shared-nothing like
exadata storage.. your exadata storage is a shared-nothing
architecture... now as you can see oracle starting to understand and
reliaze how shared-nothing architecture scales...
It is not shared-nothing, because it still is fronted by RAC.
ill cut and paste the exadata white paper here...
...
Post by fooler mail
take note that exadata is based on a massively parallel architecture
(MPP)... MPP is actually a shared-nothing architecture..
RAC is shared-all, but you add Exadata storage nodes to an existing
RAC instance. The Exadata storage nodes don't exist in a vacuum, they
are fronted by a non-MPP engine.
Post by fooler mail
but you dont realize the one who built exadata is a rocket
scientist... the same thing you dont need to build a database program
from scratch as another rocket scientist already built it for you..
you just use it...
Well I must admit that I hadn't looked at MySQL Cluster in a while,
and the last time I looked at it, it was quite crude and required a
lot of hackery (i.e. requires being a rocket scientist).

So as-of-5.0 MySQL Cluster does seem usable without a degree in rocket
science. I am curious to see though what sort of data set sizes and
transactional data rates are being achieved with MySQL Cluster. There
are some accounts that NDB Cluster is actually slower than MyISAM.
Post by fooler mail
i dont see any point from you here... any company or a person can
build a product to sell or for free to an ordinary user.. they just
simply create a cd installation disk to automate things... with
exadata it just a finish product...
It never is just about the finished product. It's also about support,
help with best practices, etc. I have a bad feeling about MySQL
Cluster because the last time I tried it, I got nowhere with it.
Besides, a product's limitations always come out when you're already
doing your non-trivial production workload, at which point, you'll be
much happier if there are some people whose job it is to help you and
you can yell at them.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
fooler mail
2009-03-21 12:50:38 UTC
Permalink
Post by Orlando Andico
This thread has gone in two directions at once: first, a generic
"shared-nothing" versus "shared everything" debate, and second,
comparing MySQL Cluster to other things.
Since this was originally a "Oracle to MySQL migration" thread I'll
ok.. let us try to simmer this down... im emphasizing here the
shared-everything vs shared-nothing architecture when it comes to
scalability... mysql cluster is just only one of the example of
shared-nothing as there are lots of shared-nothing technology out
there with different approach... ibm's db2 and ncr's teradata are the
leaders when it comes to shared-nothing database..
Post by Orlando Andico
Very true. Expertise doesn't come free with any technology.
is this coming from a closed source mentality? how about the open
source mentality?
Post by Orlando Andico
I would expect that MySQL Cluster is not a major player in the
TimesTen, SolidDB, Altibase to name a few.
but your question is who uses mysql cluster and alcatel and vodafone
are examples the ones using it... panterra networks for their on
demand unified communications.. bredbandsbolaget a broadband internet
provider in sweden for their subscriber profiles and authentication..
viasuisse AG for their real-time traffic information and others...
Post by Orlando Andico
To be honest, the last time I checked MySQL Cluster was 3 years ago
and I was not impressed. The feature set seems to be improving nicely.
I think it's time I revisit MySQL Cluster. (for which I have fooler to
thank...)
orly same true also with oracle 8 and 9 clustering.. oracle suffered
some problems too thus the improvements and address those problems
come into oracle 10...
Post by Orlando Andico
This is fine if the workload can be cleanly partitioned. However many
transactional workloads cannot be sliced-and-diced so easily.
Having shared-nothing and individual buffer caches means you cannot
guarantee transactional integrity across all members. For things like
Google (which is bulk write from robot + many many reads) basically
any read-mostly workload, this will work fine.
the industry standard for complex business intelligence processing is
the TPC-H benchmark... large-scale and multi-terabyte procesing shows
that databases using shared-nothing architecture are the ones heavily
used in this arena... furthermore.. shared-nothing architecture have
been shown to produce tremendous scaling results for OLTP application
(aside from database) when partition..
Post by Orlando Andico
As N gets large, RAC scalability for writes gets to the point of
diminishing returns. Not a limitation unique to RAC, even SMP boxes
suffer from this.
simply because shared-everything is a SMP while shared-nothing is a MPP...
Post by Orlando Andico
Shared-nothing is a known scalable solution, almost all the Top500
supercomputers are MPP. But shared-nothing has been traditionally very
hard use get working properly for traditional workloads.
TPC-H will show you the reality... see my last paragraph below...
Post by Orlando Andico
Greenplum and Netezza (and DATAllegro) have tiny market share. The
much-derided RAC has over 40% of the worldwide data warehousing
market. So if you went for these products, you'd have a much smaller
community to turn to for help. And these products (well even Exadata)
aren't designed for OLTP workloads which I believe are most
interesting to PLUG readers.
naahh... you are talking as salesman as well as a marketing guy here..
we should focus the technical merits of a shared-everything versus
shared nothing...
Post by Orlando Andico
Besides it's built for a query workload where it's permissible to lose
data. Which is an important use case but not one that most of us can
use.
thats your perception as you simply show that you dont know the inner
workings of a shared-nothing architecture... shared-nothing
architecture is reliable too..
Post by Orlando Andico
It never is just about the finished product. It's also about support,
help with best practices, etc. I have a bad feeling about MySQL
Cluster because the last time I tried it, I got nowhere with it.
Besides, a product's limitations always come out when you're already
doing your non-trivial production workload, at which point, you'll be
much happier if there are some people whose job it is to help you and
you can yell at them.
ok then.. assuming you have the million bucks money.. you are not
working from oracle and looking for a scalable database... ill refer
to you the TPC-H benchmark website:

http://www.tpc.org/tpch/

which database clustering solution you would choose as a technical guy
as usually we technical guys are always looking for price over
performance?

fooler.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-21 14:36:09 UTC
Permalink
On Sat, Mar 21, 2009 at 8:50 PM, fooler mail <fooler.mail-***@public.gmane.org> wrote:
..
Post by fooler mail
ibm's db2 and ncr's teradata are the
leaders when it comes to shared-nothing database..
Teradata isn't used for transactional at all..

And DB2 hasn't won on any SAP ERP benchmarks recently. So
shared-nothing is not the end-all solution. I did point out earlier
that for business intelligence workloads (which is what TPC-H is)
Teradata is very dominant in this space.
Post by fooler mail
Post by Orlando Andico
Very true. Expertise doesn't come free with any technology.
is this coming from a closed source mentality? how about the open
source mentality?
No, it works both ways. I don't think MySQL Cluster can be tuned
effectively by someone like... me. It takes a lot of effort to tune
any technology the best way, and that costs time and/or money. If
expertise was easy to come by, we would not be on this mailing list.


..
Post by fooler mail
but your question is who uses mysql cluster and alcatel and vodafone
are examples the ones using it... panterra networks for their on
demand unified communications.. bredbandsbolaget a broadband internet
provider in sweden for their subscriber profiles and authentication..
viasuisse AG for their real-time traffic information and others...
amusingly, all of these are from MySQL Cluster's wikipedia entry....
but I digress.


..
Post by fooler mail
the industry standard for complex business intelligence processing is
the TPC-H benchmark... large-scale and multi-terabyte procesing shows
that databases using  shared-nothing architecture are the ones heavily
used in this arena... furthermore.. shared-nothing architecture have
been shown to produce tremendous scaling results for OLTP application
(aside from database) when partition..
Yes but TPC-H is a data warehousing workload. Which is not transactional.

None of the major OLTP vendors releases TPC-C workload results
anymore, but IBM and Oracle do release SAP ERP benchmarks.
Post by fooler mail
Post by Orlando Andico
As N gets large, RAC scalability for writes gets to the point of
diminishing returns. Not a limitation unique to RAC, even SMP boxes
suffer from this.
simply because shared-everything is a SMP while shared-nothing is a MPP...
And yes... for the commodity market, everybody is still going with
SMP. My point is that MPP is either exotic or expensive. Expertise is
harder to come by.
Post by fooler mail
Post by Orlando Andico
Shared-nothing is a known scalable solution, almost all the Top500
supercomputers are MPP. But shared-nothing has been traditionally very
hard use get working properly for traditional workloads.
TPC-H will show you the reality... see my last paragraph below...
TPC-H is not a transactional workload, which is what most people do.
Post by fooler mail
Post by Orlando Andico
Greenplum and Netezza (and DATAllegro) have tiny market share. The
much-derided RAC has over 40% of the worldwide data warehousing
market. So if you went for these products, you'd have a much smaller
community to turn to for help. And these products (well even Exadata)
aren't designed for OLTP workloads which I believe are most
interesting to PLUG readers.
naahh... you are talking as salesman as well as a marketing guy here..
we should focus the technical merits of a shared-everything versus
shared nothing...
See, this is where things get to religion.

Shared-nothing is a well-known scalable technology, but for the vast
majority of the market, it's still exotic.

Top500 doesn't count. Teradata users don't count (they have lots of
money). Core banking systems, even our beloved web forums, mail
servers, most of them run on shared-everything. Shared-nothing is
still not within the reach of the "common technical user."

Which is my entire point -- shared-nothing requires more rocket
science skills than shared-everything. Of course we all know the
limitations and performance plateau of shared-everything: but that
plateau is still high enough for most use cases.


..
Post by fooler mail
Post by Orlando Andico
Besides it's built for a query workload where it's permissible to lose
data. Which is an important use case but not one that most of us can
use.
thats your perception as you simply show that you dont know the inner
workings of a shared-nothing architecture... shared-nothing
architecture is reliable too..
I was talking about Google's use case being query-heavy, and tolerant
of data loss. Not shared-nothing in general.

Off topic but one of the products I've been working with --- Oracle
Coherence --- is a distributed shared-nothing architecture. So I'm not
exactly ignorant of "the workings of a shared-nothing architecture."


..
Post by fooler mail
ok then..  assuming you have the million bucks money.. you are not
working from oracle and looking for a scalable database... ill refer
http://www.tpc.org/tpch/
which database clustering solution you would choose as a technical guy
as usually we technical guys are always looking for price over
performance?
I would use Exasol for TPC-H. No brainer. 1/25th the price of Oracle
solution and stratospheric performance.

But TPC-H is not transactional and is hardly something that is
relevant to PLUG readers.

Most BI/DW workloads simply suck at transactional loads. But they
don't have to be good at transactional loads because complex
(read-only) query performance is paramount.

But again, that's not something we on PLUG generally do as part of our
daily lives.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
fooler mail
2009-03-21 18:02:17 UTC
Permalink
On Sat, Mar 21, 2009 at 10:36 PM, Orlando Andico <orly.andico-***@public.gmane.org> wrote:
now we are dealing another topic to debate but ill answer you...
Post by Orlando Andico
Teradata isn't used for transactional at all..
teradata database specializing in data warehousing but their database
is capable of transactional too.. it can do transaction rollback (eg.
begin transaction (BT) and end transaction (ET) commands)...

look at their customers section at wikipedia...

http://en.wikipedia.org/wiki/Teradata

bank of america.. you know how highly critical monetary transactions
are there...
Post by Orlando Andico
And DB2 hasn't won on any SAP ERP benchmarks recently. So
shared-nothing is not the end-all solution. I did point out earlier
that for business intelligence workloads (which is what TPC-H is)
Teradata is very dominant in this space.
as what i have said in my first post.. there are pros and cons for
everything.. know your workload first and choose the right database
for you.. as you can see.. mysql has its own place based on the TPC
benchmark...
Post by Orlando Andico
No, it works both ways. I don't think MySQL Cluster can be tuned
effectively by someone like... me. It takes a lot of effort to tune
any technology the best way, and that costs time and/or money. If
expertise was easy to come by, we would not be on this mailing list.
tuning is a black art...
Post by Orlando Andico
amusingly, all of these are from MySQL Cluster's wikipedia entry....
but I digress.
but that proves that they used mysql clustering and they are happy
with it without a problem on their side...
Post by Orlando Andico
And yes... for the commodity market, everybody is still going with
SMP. My point is that MPP is either exotic or expensive. Expertise is
harder to come by.
the argument here is scalability... when SMP cant scale.. the next
step is MPP..
Post by Orlando Andico
Shared-nothing is a well-known scalable technology, but for the vast
majority of the market, it's still exotic.
exotic is meaningless if you need to scale as your workload grows...
Post by Orlando Andico
Top500 doesn't count. Teradata users don't count (they have lots of
money). Core banking systems, even our beloved web forums, mail
servers, most of them run on shared-everything. Shared-nothing is
still not within the reach of the "common technical user."
Which is my entire point -- shared-nothing requires more rocket
science skills than shared-everything. Of course we all know the
limitations and performance plateau of shared-everything: but that
plateau is still high enough for most use cases.
orly because your mindset wont adapt to the next level...your mindset
is just like an old people that dont want to touch that computer or
cellphone as it is hard for them to learn...
Post by Orlando Andico
But again, that's not something we on PLUG generally do as part of our
daily lives.
but someday they do and others are on it...

fooler.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-22 02:00:53 UTC
Permalink
This discussion is rapidly becoming irrelevant... we were originally
talking about MySQL and now see where we've ended up...
Post by fooler mail
now we are dealing another topic to debate but ill answer you...
Post by Orlando Andico
Teradata isn't used for transactional at all..
teradata database specializing in data warehousing but their database
is capable of transactional too.. it can do transaction rollback (eg.
begin transaction (BT) and end transaction (ET) commands)...
look at their customers section at wikipedia...
http://en.wikipedia.org/wiki/Teradata
bank of america.. you know how highly critical monetary transactions
are there...
So we're back to quoting Wikipedia.

The existence of a transactional feature does not mean that it works
well. You can do transactions in SybaseIQ for example but they are
painfully slow.

Most banks do have data warehousing loads, so just because BofA uses
Teradata doesn't mean they use it for transactional loads.

There's not much point really bandying Wikipedia-based arguments
around. From my own personal knowledge, which isn't that vast I admit,

..
Post by fooler mail
Post by Orlando Andico
amusingly, all of these are from MySQL Cluster's wikipedia entry....
but I digress.
but that proves that they used mysql clustering and they are happy
with it without a problem on their side...
No, you are reading too much into a Wikipedia entry.

Not all references are "happy" and don't have problems.
Post by fooler mail
Post by Orlando Andico
And yes... for the commodity market, everybody is still going with
SMP. My point is that MPP is either exotic or expensive. Expertise is
harder to come by.
the argument here is scalability... when SMP cant scale.. the next
step is MPP..
No, my argument was that SMP (or shared-all) can scale, to a certain
extent. And the extent of that scaling still exceeds most use case
requirements.
Post by fooler mail
Post by Orlando Andico
Shared-nothing is a well-known scalable technology, but for the vast
majority of the market, it's still exotic.
exotic is meaningless if you need to scale as your workload grows...
Yeah... but I personally know of no use cases that absolutely require
MPP-class scaling. Of course there will always be some use case which
requires massive scaling. On all the things I've worked on, I've only
seen one which requires more scaling than traditional shared-all can
provide.



..
Post by fooler mail
Post by Orlando Andico
Which is my entire point -- shared-nothing requires more rocket
science skills than shared-everything. Of course we all know the
limitations and performance plateau of shared-everything: but that
plateau is still high enough for most use cases.
orly because your mindset wont adapt to the next level...your mindset
is just like an old people that dont want to touch that computer or
cellphone as it is hard for them to learn...
Oh, so I've got an obsolete mindset now.

Please go look up Oracle Coherence. I've been workign with it for a
couple years now. Unfortunately most people DO have a conventional
mindset and nobody's deploying this solution in ASEAN.


..
Post by fooler mail
Post by Orlando Andico
But again, that's not something we on PLUG generally do as part of our
daily lives.
but someday they do and others are on it...
True. But it's irrelevant to the original topic of this discussion
anyway. Saying "others are on it" is like saying "some people have a
Benz S500." It's irrelevant for most people.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
fooler mail
2009-03-22 03:47:36 UTC
Permalink
Post by Orlando Andico
This discussion is rapidly becoming irrelevant... we were originally
talking about MySQL and now see where we've ended up...
yes i know and im trying to simmer it down as you injecting more
irrelevant topic as i focus more on scalability between
shared-everything vs shared-nothing in technical terms as what i
started with my first post... of course few are using it simply
because few are having enormous workloads...
Post by Orlando Andico
Teradata doesn't mean they use it for transactional loads.
your original argument was teradata was not used for transaction at
*all*.. but i prove it was capable for transactional loads too.. but
you keep on injecting more irrelevant topic that this debate keeps on
longer..
Post by Orlando Andico
No, you are reading too much into a Wikipedia entry.
naahhh.. wikipedia is there for reference just like when your doing
your thesis you need reference too.. your original argument who *uses*
mysql clustering.. and i prove those companies are using it... but you
keep on injecting another irrelevant topic to keep this debate
longer..
Post by Orlando Andico
No, my argument was that SMP (or shared-all) can scale, to a certain
extent. And the extent of that scaling still exceeds most use case
requirements.
and when time comes your requirements doesnt meet the smp scalability?
as what i pointed out in my first post.. the next logical thing to do
is to go mpp...

this is the same true when the processing power of a cpu reaches to an
atomic level.. you cant further improve that.. the next step is to go
quantum processing...
Post by Orlando Andico
Oh, so I've got an obsolete mindset now.
Please go look up Oracle Coherence. I've been workign with it for a
couple years now.
oracle acquired tangosol product in 2007 and dubbed it as oracle
coherence.. coherence is a shared infrastructure.. it is an
application server-level caching product.. sharing data between nodes
in a cluster via ram cache for frequently used data.. it bypass oracle
data buffers by sharing at the application server ... it is one of
your solution that RAC for certain applications..
Post by Orlando Andico
True. But it's irrelevant to the original topic of this discussion
anyway. Saying "others are on it" is like saying "some people have a
Benz S500." It's irrelevant for most people.
see.. now we are getting into what joebert tried to deliver here..

fooler.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-22 04:57:54 UTC
Permalink
Post by fooler mail
Post by Orlando Andico
This discussion is rapidly becoming irrelevant... we were originally
talking about MySQL and now see where we've ended up...
yes i know and im trying to simmer it down as you injecting more
irrelevant topic as i focus more on scalability between
shared-everything vs shared-nothing in technical terms as what i
started with my first post... of course few are using it simply
because few are having enormous workloads...
This topic wasn't originally between shared-everything and shared-nothing.

..
Post by fooler mail
Post by Orlando Andico
No, you are reading too much into a Wikipedia entry.
naahhh.. wikipedia is there for reference just like when your doing
your thesis you need reference too.. your original argument who *uses*
mysql clustering.. and i prove those companies are using it... but you
keep on injecting another irrelevant topic to keep this debate
longer..
No, you say they use it and they are happy and they have no problems.
That's quite a stretch from the Wikipedia entry, that's far more than
just referencing.

Had you simply said "these companies use it" then that would be fine.

..
Post by fooler mail
oracle acquired tangosol product in 2007 and dubbed it as oracle
coherence..  coherence is a shared infrastructure.. it is an
application server-level caching product.. sharing data between nodes
in a cluster via ram cache for frequently used data.. it bypass oracle
data buffers by sharing at the application server ... it is one of
your solution that RAC for certain applications..
My point is that you broadly labeling me as unable to understand
shared-everything is erroneous. And your understanding of Oracle
Coherence is very incomplete. You say I inject various topics to
prolong the debate but the reality is that you can't resist taking
personal digs at me.
Post by fooler mail
Post by Orlando Andico
True. But it's irrelevant to the original topic of this discussion
anyway. Saying "others are on it" is like saying "some people have a
Benz S500." It's irrelevant for most people.
see.. now we are getting into what joebert tried to deliver here..
The original topic was... issues you may face in migrating from MySQL
to Oracle. I initially pointed out areas where it's simply not
possible to move from MySQL to Oracle (lacking features). Then you
come in and say shared-nothing is superior... that RAC exhibits
negative scalability... that's where this prolonged debate began.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
fooler mail
2009-03-22 06:56:48 UTC
Permalink
Post by Orlando Andico
This topic wasn't originally between shared-everything and shared-nothing.
take note where i bump in... they are talking about RAC scalability...
i just emphasize between shared-everything and shared-nothing...
Post by Orlando Andico
No, you say they use it and they are happy and they have no problems.
That's quite a stretch from the Wikipedia entry, that's far more than
just referencing.
then challenge that wikipedia page those *prominent* users section
being mentioned... when i read the word *prominent*... it is logically
to think they are happy with it...
Post by Orlando Andico
My point is that you broadly labeling me as unable to understand
shared-everything is erroneous.
no it is the other way around - shared-nothing...
Post by Orlando Andico
And your understanding of Oracle
Coherence is very incomplete.
and i dont want to explain from you further... as it will keep on
meaningless debate...
Post by Orlando Andico
You say I inject various topics to
prolong the debate but the reality is that you can't resist taking
personal digs at me.
personal is nothing to do with here... but the irrelevant topics you
were injecting that prolong the debate... like this one now for
example...
Post by Orlando Andico
The original topic was... issues you may face in migrating from MySQL
to Oracle. I initially pointed out areas where it's simply not
possible to move from MySQL to Oracle (lacking features).
the previous posts of others already answered those questions...

to make it clear.. i want to migrate from oracle to mysql because
those features in mysql already serve my needs and cost me less.. no
need to explain those oracle features to me...
Post by Orlando Andico
Then you
come in and say shared-nothing is superior... that RAC exhibits
negative scalability...
because it simply true...
Post by Orlando Andico
that's where this prolonged debate began.
it is really upto you orly if you want to continue this useless debate
already...

fooler.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-22 07:19:08 UTC
Permalink
Post by fooler mail
Post by Orlando Andico
This topic wasn't originally between shared-everything and shared-nothing.
take note where i bump in... they are talking about RAC scalability...
i just emphasize between shared-everything and shared-nothing...
And RAC does scale. Not to the clouds, but for almost all practical
workloads it does.
Post by fooler mail
Post by Orlando Andico
No, you say they use it and they are happy and they have no problems.
That's quite a stretch from the Wikipedia entry, that's far more than
just referencing.
then challenge that wikipedia page those *prominent* users section
being mentioned... when i read the word *prominent*... it is logically
to think they are happy with it...
Nope that's stretching the interpretation of the Wikipedia page.

I can give some examples of high-profile public references turning
sour internally but I won't.
Post by fooler mail
Post by Orlando Andico
My point is that you broadly labeling me as unable to understand
shared-everything is erroneous.
no it is the other way around - shared-nothing...
I stand corrected -- so you do admit saying I have an old person's
mindset and cannot understand shared-nothing. Please correct me if my
interpretation is wrong.
Post by fooler mail
Post by Orlando Andico
And your understanding of Oracle
Coherence is very incomplete.
and i dont want to explain from you further... as it will keep on
meaningless debate...
Yes it is meaningless. But you can't get away with calling people out
as unable to understand something, and then dropping the topic when
YOU get called out.
Post by fooler mail
Post by Orlando Andico
You say I inject various topics to
prolong the debate but the reality is that you can't resist taking
personal digs at me.
personal is nothing to do with here... but the irrelevant topics you
were injecting that prolong the debate... like this one now for
example...
Yes perhaps it's irrelevant. But I didn't question your competence.


..
Post by fooler mail
the previous posts of others already answered those questions...
to make it clear.. i want to migrate from oracle to mysql because
those features in mysql already serve my needs and cost me less.. no
need to explain those oracle features to me...
Post by Orlando Andico
Then you
come in and say shared-nothing is superior... that RAC exhibits
negative scalability...
because it simply true...
See. That's the thing.

You cannot get away with broad generalizations like "RAC exhibits
negative scalability... and it [sic] simply true" without getting
challenged.
Post by fooler mail
Post by Orlando Andico
that's where this prolonged debate began.
it is really upto you orly if you want to continue this useless debate
already...
Well yeah this discussion has turned sour and I feel embarrassed to
the PLUG community at large. But I'm not solely at fault here.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
John Peter Loh
2009-03-22 07:24:30 UTC
Permalink
That's okay. Nosebleed to most, including me. Clustering and scaling
apps interest me though. It was fun to watch. Give it a break, guys.
It's Sunday. ;)
Post by Orlando Andico
Well yeah this discussion has turned sour and I feel embarrassed to
the PLUG community at large. But I'm not solely at fault here.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Alexander Panganiban Santiago
2009-03-22 07:38:00 UTC
Permalink
Agree,

been researching for the terms used here!

while someone here benefits from the spark of mental swords clashing!

peace!
Post by John Peter Loh
That's okay. Nosebleed to most, including me. Clustering and scaling
apps interest me though. It was fun to watch. Give it a break, guys.
It's Sunday. ;)
Post by Orlando Andico
Well yeah this discussion has turned sour and I feel embarrassed to
the PLUG community at large. But I'm not solely at fault here.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
--
Alexander Panganiban Santiago

"malawak ang mundo para magdamot"
Registered Linux User # 384399
Zak B. Elep
2009-03-22 07:46:32 UTC
Permalink
Post by John Peter Loh
That's okay. Nosebleed to most, including me. Clustering and scaling
apps interest me though. It was fun to watch. Give it a break, guys.
It's Sunday. ;)
Amen. Probably needs more frag-or-be-fragged action in Tremulous or
Sauerbraten (if you don't do Team Fortress 2 or Left4Dead...)
--
Zak B. Elep || http://zakame.spunge.org
zakame-GeWIH/nMZzLQT0dZR+***@public.gmane.org || zakame-NlwAFEu9kz0dnm+***@public.gmane.org || zakame-***@public.gmane.org
1486 7957 454D E529 E4F1 F75E 5787 B1FD FA53 851D
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
fooler mail
2009-03-22 07:54:09 UTC
Permalink
Post by Orlando Andico
And RAC does scale. Not to the clouds, but for almost all practical
workloads it does.
sell your statement that to google, walmart, amazon, bank of america,
coca cola, fed ex, etc, etc... and ill be happy for you if they
switch...
Post by Orlando Andico
Nope that's stretching the interpretation of the Wikipedia page.
I can give some examples of high-profile public references turning
sour internally but I won't.
then create your wikipedia page then... author of the page - oracle ...
Post by Orlando Andico
Yes it is meaningless. But you can't get away with calling people out
as unable to understand something, and then dropping the topic when
YOU get called out.
useless...
Post by Orlando Andico
See. That's the thing.
You cannot get away with broad generalizations like "RAC exhibits
negative scalability... and it [sic] simply true" without getting
challenged.
already explained in my previous posts and another useless debate...
Post by Orlando Andico
Well yeah this discussion has turned sour and I feel embarrassed to
the PLUG community at large. But I'm not solely at fault here.
well to make you happy.. both of us are at fault here.. we ended up
in a useless debate...

fooler.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-22 09:57:04 UTC
Permalink
Well, in the hopes that *some* good comes out of all this....

1) for most production environments that are already on Oracle,
switching to MySQL can range from straightforward (e.g. you are just
using Hibernate) to impossible (if you're using a specific feature
like DICOM which is not present in MySQL)


2) shared-nothing architectures scale well but I believe they continue
to be on the margins. Fooler conveniently points out that a lot of
large enterprises use shared-nothing: but all of these enterprises
have something in common, lots of money and lots of in-house
expertise.

My original contention stands: shared-nothing is not as
straightforward although it scales higher.

3) RAC scales well while the number of nodes is small (e.g. < 10).
Generalizing that "RAC scales negatively" is inaccurate and a
dis-service to all the people using RAC to run their business.

4) MySQL Cluster seems to have matured quite a bit in the last couple
years and Sun is really pushing it in the telecoms space. Although to
be honest I have never gone up against MySQL in a telecom account,
more common is Solid and Altibase.


I would not push Oracle solutions where they are not appropriate; in
fact if I was setting up my own startup or something I would certainly
use MySQL (clustered or non-clustered) to keep my burn rate down while
I was ramping up.

All technologies have their drawbacks, but you have to balance budget
versus development time versus market acceptability. I think it is
safe to say that outside Web 2.0 applications MySQL hasn't gotten that
much traction.



As an aside, Fooler loves to point out that I'm biased and an Oracle
shill. But that's OK, everybody here knows where I come from and I
make no effort to conceal my affiliations.

I do wonder why he seems overly bitter about his Oracle and RAC
experience (please see his original posts) and interestingly enough
all these years on PLUG I don't know Fooler's real name. So he can
call out people and brand them as "useless" with impunity.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-22 11:38:34 UTC
Permalink
http://3.bp.blogspot.com/_0hNQ3clcRl4/SOFqXk_1onI/AAAAAAAAAPI/6aTn9-3TMnM/s1600-h/orcl.bmp
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-20 11:10:07 UTC
Permalink
On Fri, Mar 20, 2009 at 1:53 PM, fooler mail <fooler.mail-***@public.gmane.org> wrote:
..
Post by fooler mail
shared-nothing architecture where each node is independent and
self-sufficient and there is no single point of contention across the
system... one good example and demonstrated how effective
shared-nothing architecture and scale almost *infinitely* by adding an
inexpensive node since there is no single bottleneck to slow down the
system is google database (using a flat file spreads across the
system)... amazon is doing this also... with a massive read and write
transactions.. shared-nothing is the logical choice here...
i have experienced with oracle rac (10gR2) and experienced how it was
bog down the whole RAC system too.. tsk tsk tsk
1) has anyone really looked into MySQL Cluster?

I have. That shared-nothing implementation is.. really hard to use in
practice. You are limited to available memory (!) because NDB Cluster
is an in-memory implementation. And look at their references.. who has
heard of Zillow.com?

I looked into MySQL Cluster before at a prior role, and I concluded
that for what we wanted, MySQL Cluster was more trouble than it was
worth. My understanding is that due to the replication topology, your
writes will not scale well as the number of nodes increases as you
have to propagate changes OVER THE NETWORK to the other nodes.

So much for linear scalability...

MySQL Cluster's overview claims "few milliseconds" query latency and
"tens of thousands" of transactions per second. I'd bet those are
mostly reads.

Oracle In-Memory Database can do SIX MICROSECOND selects and FIFTEEN
MICROSECOND updates, and up to 750K selects/second (write performance
is considerably less, around 200K/second). Granted this is not Oracle
DB, but it's the more logical comparison to MySQL Cluster.


2) Amazon has the world's largest data warehouse. And it's not running
a shared-nothing MySQL architecture.
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
fooler mail
2009-03-20 14:41:12 UTC
Permalink
Post by Orlando Andico
1) has anyone really looked into MySQL Cluster?
I have. That shared-nothing implementation is.. really hard to use in
practice. You are limited to available memory (!) because NDB Cluster
is an in-memory implementation. And look at their references.. who has
heard of Zillow.com?
if you said about rac that people dont know how to tune or configure
rac.. they same true also to people who dont know how to configure
mysql cluster...

alcatel uses mysql cluster for their home location register (HLR)..
HLR is a central database that contains details of each mobile phone
subscriber that is authorized to use the gsm network.. vodafone in UK
uses mysql cluster too..
Post by Orlando Andico
I looked into MySQL Cluster before at a prior role, and I concluded
that for what we wanted, MySQL Cluster was more trouble than it was
worth. My understanding is that due to the replication topology, your
writes will not scale well as the number of nodes increases as you
have to propagate changes OVER THE NETWORK to the other nodes.
So much for linear scalability...
the mysql cluster replication provides highly availability in case one
node fails the other node will take care of that data... usually
replication took place when an update on the master server... it
propagates to a slave server and not to the rest of the nodes...
Post by Orlando Andico
MySQL Cluster's overview claims "few milliseconds" query latency and
"tens of thousands" of transactions per second. I'd bet those are
mostly reads.
Oracle In-Memory Database can do SIX MICROSECOND selects and FIFTEEN
MICROSECOND updates, and up to 750K selects/second (write performance
is considerably less, around 200K/second). Granted this is not Oracle
DB, but it's the more logical comparison to MySQL Cluster.
2) Amazon has the world's largest data warehouse. And it's not running
a shared-nothing MySQL architecture.
i didnt mention amazon uses mysql ... what i mentioned was that amazon
used a shared-nothing architecture also just like google for
scalability reason as oracle rac was well known of slow i/o
performance (eg. weaknesses in parallel query planning.. inefficient
movement of data between nodes for execution of complex joins and
aggregations) thus exadata storage exists...

fooler.
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Orlando Andico
2009-03-20 03:05:09 UTC
Permalink
Oracle Database Express Edition is free for deployment as well.

But there is no support (you'd have to go to the Oracle Forums and ask around).

Also it is limited to 1 core, 1GB of RAM, and 4GB of data.

IBM DB2 Express-C is a much less crippled offering, up to 2 CPUs,
unlimited data space. However both Oracle and DB2 are large and
daunting offerings, and going with DB2 just because "I can get
enterprise database for free beer" is a bad justification. Not to
mention the DB2 community is smaller than the Oracle community, so
less help is forthcoming unless you're a DB2 expert already.

You should use the technology that meets your needs and budget.

Deriding people who use open source as poor losers is un-productive.
Defending open-source as the end-all and portraying closed-source
users as ignorant folks with lots of money is wrong as well.

I'm sure everyone here would gladly ditch MySQL for Oracle for their
usual deployments, if money was not an issue. I'd gladly ditch my
Mazda for a Benz if money was not an issue. There is a very definite
technical advantage to Oracle, and anyone who says otherwise is
mistaken.



On Thu, Mar 19, 2009 at 6:46 PM, Kelsey Hartigan Go
Post by Kelsey Hartigan Go
There is a free version of Oracle for development...but severely
restricted...similar to microsoft sqlexpress.
and Db2express (but db2 and sqlexpress you can use for production, oracle,
you can't)
--
Orlando Andico
+63.2.976.8659 | +63.920.903.0335
_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph
Loading...