Re: PostgreSQL 9.0 x64 bit pgbench TPC very low question?

From: Scott Marlowe <scott(dot)marlowe(at)gmail(dot)com>
To: tuanhoanganh <hatuan05(at)gmail(dot)com>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: PostgreSQL 9.0 x64 bit pgbench TPC very low question?
Date: 2010-12-20 14:21:40
Message-ID: AANLkTikF2dNay2Dbih4KVGskM07PMzwP3btyBj25mbwX@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Mon, Dec 20, 2010 at 7:10 AM, tuanhoanganh <hatuan05(at)gmail(dot)com> wrote:
> Here is my new pgbench's point
>
> pgbench -h 127.0.0.1 -p 9999 -U postgres -c 200 -t 100 -s 10 pgbench

Your -c should always be the same or lower than -s. Anything higher
and you're just thrashing your IO system waiting for locks. Note that
-s is ignored on runs if you're not doing -i. Also -t 100 is too
small to get a good test, try at least 1000 or 10000 and let it run a
minute.

> Scale option ignored, using pgbench_branches table count = 10
> starting vacuum...end.
> transaction type: TPC-B (sort of)
> scaling factor: 10
> query mode: simple
> number of clients: 200
> number of threads: 1
> number of transactions per client: 100
> number of transactions actually processed: 20000/20000
> tps = 202.556936 (including connections establishing)
> tps = 225.498811 (excluding connections establishing)

> I have server computer install Windows 2008R2, PostgreSQL 9.0.1 64 bit, 8G
> RAM, RAID 10 - 4 disks
>
> Is it common pgbench 's point with my server ?

That's a pretty reasonable number for that class machine. I assume
you do NOT have a battery backed caching RAID controller or it would
be WAY higher.

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Scott Marlowe 2010-12-20 14:24:14 Re: PostgreSQL 9.0 x64 bit pgbench TPC very low question?
Previous Message tuanhoanganh 2010-12-20 14:10:58 Re: PostgreSQL 9.0 x64 bit pgbench TPC very low question?