Re: [PERFORM] OSDL Database Test Suite 3 is available on PostgreSQL

2003-08-04 Thread Shridhar Daithankar
On 4 Aug 2003 at 15:33, Manfred Koizar wrote:

> On 01 Aug 2003 11:04:10 -0700, Jenny Zhang <[EMAIL PROTECTED]> wrote:
> >A sample OSDL-DBT3 test result report can be found at:
> >http://khack.osdl.org/stp/276912/
> >
> >Your comments are welcome,

I could not get postgresql .conf so I will combine the comments.

1. Effective cache size, already mentioned
2. Sort memory already mentioned.
3. Was WAL put on different drive?
4. Can you try with autovacuum daemon and 7.4beta when it comes out..
5. What was the file system? Ext2/Ext3/reiser/XFS?



Is there any comparison available for other databases.. Could be interesting to 
see..:-)



Thanks for the good work. I understand it must have been quite an effort to run 
it..

Keep it up..

Bye
 Shridhar

--
Fourth Law of Revision: It is usually impractical to worry beforehand about 
interferences -- if you have none, someone will make one for you.


---(end of broadcast)---
TIP 9: the planner will ignore your desire to choose an index scan if your
  joining column's datatypes do not match


Re: [PERFORM] OSDL Database Test Suite 3 is available on PostgreSQL

2003-08-04 Thread scott.marlowe
On Mon, 4 Aug 2003, Manfred Koizar wrote:

> On 01 Aug 2003 11:04:10 -0700, Jenny Zhang <[EMAIL PROTECTED]> wrote:
> >A sample OSDL-DBT3 test result report can be found at:
> >http://khack.osdl.org/stp/276912/
> >
> >Your comments are welcome,
> 
> | effective_cache_size   | 1000
> 
> With 4GB of memory this is definitely too low and *can* (note that I
> don't say *must*) lead the planner to wrong decisions.
> 
> | shared_buffers | 15200
> 
> ... looks reasonable.  Did you test with other values?
> 
> | sort_mem   | 524288
> 
> This is a bit high, IMHO, but might be ok given that DBT3 is not run
> with many concurrent sessions (right?).
> http://khack.osdl.org/stp/276912/results/plot/vmstat_swap.png shows
> some swapping activity towards the end of the run which could be
> caused by a too high sort_mem setting.

Sorry, my last email shot off out of the gun before it was completed...

To repeat:  Don't forget to set effective_cache_size the same way as 
shared buffers (i.e. it's in 8k blocks for most systems.)  If you have a 
machine with 4 gig ram, and 3 gigs is available as disk cache, then 
divide out 3 gigs by 8k to get the right number.  My quick calculation 
shows that being about 393216 blocks.


---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]


Re: [PERFORM] OSDL Database Test Suite 3 is available on PostgreSQL

2003-08-04 Thread scott.marlowe
On Mon, 4 Aug 2003, Manfred Koizar wrote:

> On 01 Aug 2003 11:04:10 -0700, Jenny Zhang <[EMAIL PROTECTED]> wrote:
> >A sample OSDL-DBT3 test result report can be found at:
> >http://khack.osdl.org/stp/276912/
> >
> >Your comments are welcome,
> 
> | effective_cache_size   | 1000
> 
> With 4GB of memory this is definitely too low and *can* (note that I
> don't say *must*) lead the planner to wrong decisions.
> 
> | shared_buffers | 15200
> 
> ... looks reasonable.  Did you test with other values?
> 
> | sort_mem   | 524288
> 
> This is a bit high, IMHO, but might be ok given that DBT3 is not run
> with many concurrent sessions (right?).
> http://khack.osdl.org/stp/276912/results/plot/vmstat_swap.png shows
> some swapping activity towards the end of the run which could be
> caused by a too high sort_mem setting.

And, as always, don't forget to set effect_cache_size.


---(end of broadcast)---
TIP 7: don't forget to increase your free space map settings


Re: [PERFORM] OSDL Database Test Suite 3 is available on PostgreSQL

2003-08-04 Thread Manfred Koizar
On 01 Aug 2003 11:04:10 -0700, Jenny Zhang <[EMAIL PROTECTED]> wrote:
>A sample OSDL-DBT3 test result report can be found at:
>http://khack.osdl.org/stp/276912/
>
>Your comments are welcome,

| effective_cache_size   | 1000

With 4GB of memory this is definitely too low and *can* (note that I
don't say *must*) lead the planner to wrong decisions.

| shared_buffers | 15200

... looks reasonable.  Did you test with other values?

| sort_mem   | 524288

This is a bit high, IMHO, but might be ok given that DBT3 is not run
with many concurrent sessions (right?).
http://khack.osdl.org/stp/276912/results/plot/vmstat_swap.png shows
some swapping activity towards the end of the run which could be
caused by a too high sort_mem setting.

Servus
 Manfred

---(end of broadcast)---
TIP 6: Have you searched our list archives?

   http://archives.postgresql.org


Re: [PERFORM] OSDL Database Test Suite 3 is available on PostgreSQL

2003-08-01 Thread Josh Berkus
Jenny,

> OSDL has ported OSDL Database Test Suite 3(OSDL-DBT3) to PostgreSQL.
> It drives the database with an ad-hoc decision support workload.  It
> helps database developers and users to identify database performance
> issues.

Way, way, cool!   We've been waiting for something like this eagerly.   I 
really look forward to trying it out.

Further, OSDL-DBT3 can hopefully serve as the kernel of our internal 
performance option testing suite.  We'll evaluate very soon.

Thanks so much for your hard work!

-- 
Josh Berkus
PostgreSQL Core Team
San Francisco

---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster


[PERFORM] OSDL Database Test Suite 3 is available on PostgreSQL

2003-08-01 Thread Jenny Zhang
OSDL has ported OSDL Database Test Suite 3(OSDL-DBT3) to PostgreSQL.  
It drives the database with an ad-hoc decision support workload.  It
helps database developers and users to identify database performance
issues.

OSDL-DBT3 is derived from TPC-H benchmark.  TPC-H is an ad-hoc decision
support benchmark.  It has 22 complicated queries doing a lot of
table/index scanning, sorting and grouping by.  Details about TPC-H can
be found at: http://www.tpc.org/tpch/

Though OSDL-DBT3 is based on TPC-H, it deviates from TPC-H
significantly.  It skipped many of the requirements for audit purpose,
as well as added flexibility.  OSDL-DBT3 performance test results are
not comparable to TPC-H results.

OSDL-DBT3 tarball can be downloaded from:
http://sourceforge.net/projects/osdldbt/
The source can be downloaded from source forge cvs tree and osdl bk tree
http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/osdldbt/dbt3/
bk://developer.osdl.org/dbt3

OSDL-DBT3 is also implemented on OSDL Scalable Test Platform(STP).  The
platform provides a framework where users run a set of performance and
scalability tests on various hardware platforms.  Currently, OSDL-DBT3
is running against PostgreSQL 7.3.3 only.  We plan to improve it so that
it can run against PostgreSQL patches.  To find more information about
STP, visit: http://www.osdl.org/stp/.  

A sample OSDL-DBT3 test result report can be found at:
http://khack.osdl.org/stp/276912/

Your comments are welcome,
Regards,
Jenny
-- 
Jenny Zhang
Open Source Development Lab Inc 
12725 SW Millikan Way
Suite 400
Beaverton, OR 97005
(503)626-2455 ext 31


---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]