Обсуждение: JDBC performance issue

Поиск
Список
Период
Сортировка

JDBC performance issue

От
CS DBA
Дата:
Hi All;

We have a client running Pentaho to migrate data.

They recently upgraded Pentaho which upgraded their JDBC driver from 8.4
(postgresql-8.4-703.jdbc4.jar) to 9.1 (postgresql-8.4-703.jdbc4.jar).
They have a test set which updates 1000 rows, with the old driver it
takes about 30 seconds, but the new driver takes 3x longer.

This is on PostgreSQL 9.1

Thoughts?

Thanks in advance...





Re: JDBC performance issue

От
Dave Cramer
Дата:
You said they upgraded the driver but the versions you stated are the same ???

Dave Cramer

dave.cramer(at)credativ(dot)ca
http://www.credativ.ca


On Wed, Feb 5, 2014 at 12:23 PM, CS DBA <cs_dba@consistentstate.com> wrote:
Hi All;

We have a client running Pentaho to migrate data.

They recently upgraded Pentaho which upgraded their JDBC driver from 8.4 (postgresql-8.4-703.jdbc4.jar) to 9.1 (postgresql-8.4-703.jdbc4.jar).  They have a test set which updates 1000 rows, with the old driver it takes about 30 seconds, but the new driver takes 3x longer.

This is on PostgreSQL 9.1

Thoughts?

Thanks in advance...





--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: JDBC performance issue

От
David Johnston
Дата:
CS DBA wrote
> Hi All;
>
> We have a client running Pentaho to migrate data.
>
> They recently upgraded Pentaho which upgraded their JDBC driver from 8.4
> (postgresql-8.4-703.jdbc4.jar) to 9.1 (postgresql-8.4-703.jdbc4.jar).
> They have a test set which updates 1000 rows, with the old driver it
> takes about 30 seconds, but the new driver takes 3x longer.
>
> This is on PostgreSQL 9.1
>
> Thoughts?

Good News!  This seems to be a re-producible problem.

Bad News!  Either you/your client will need to discover the bottle-neck or
you will need to provide the test set/routine to the public so someone else
can try.

Also, maybe run the test set against and with 9.2 and/or 9.3 to see if the
problem still exists there...

Note, I presume you mean "updates 1000s of rows" (i.e., considerably more
than 1,000) - 30s to update exactly 1000 rows doesn't seem that impressive
but then again no clue what is involved so maybe that is good.  Note that
means it is unclear if the "UPDATE" phase or some required "SELECT" is
mis-behaving which is a critical detail.

David J.




--
View this message in context: http://postgresql.1045698.n5.nabble.com/JDBC-performance-issue-tp5790710p5790714.html
Sent from the PostgreSQL - general mailing list archive at Nabble.com.