Re: PostgreSQL 8.4 performance tuning questions

From: Arjen van der Meijden <acmmailing(at)tweakers(dot)net>
To: Scott Carey <scott(at)richrelevance(dot)com>
Cc: Kevin Grittner <Kevin(dot)Grittner(at)wicourts(dot)gov>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Matthew Wakeling <matthew(at)flymine(dot)org>, "pgsql-performance(at)postgresql(dot)org" <pgsql-performance(at)postgresql(dot)org>
Subject: Re: PostgreSQL 8.4 performance tuning questions
Date: 2009-07-30 19:56:02
Message-ID: 4A71FAD2.9020607@tweakers.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On 30-7-2009 20:46 Scott Carey wrote:
> Of course Compression has a HUGE effect if your I/O system is half-decent.
> Max GZIP compression speed with the newest Intel CPU's is something like
> 50MB/sec (it is data dependant, obviously -- it is usually closer to
> 30MB/sec). Max gzip decompression ranges from 50 to 150MB/sec (it can get
> really high only if the ratio is extremely large, like if you compress a
> repeating sequence of 256 bytes).

I just ran some quick numbers on our lightly loaded Nehalem X5570 (2.93+
Ghz depending on turbo-mode). I compressed a 192MB text file I had at
hand using gzip -1, -2, -3, -6 and -9 and outputted its results to
/dev/null. The file was in the kernels file cache all the time and I did
the tests 3 times.

Gzip -1 reached 54MB/s, -2 got 47MB/s, -3 got 32MB/s, -6 got 18MB/s and
-9 got to 12MB/s. Just running cat on the file made it do 6400MB/s (i.e.
it took 0.030 seconds to copy the file from memory to nowhere).
Those files where respectively 69MB, 66MB, 64MB, 59MB and 58MB.

Gunzip on the -1 file took 1.66 seconds, i.e. it read data at 41MB/s and
outputted it to /dev/null at 115MB/s. The -9 file took 1.46s, so it read
40MB/s and wrote 131MB/s.

Best regards,

Arjen

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Scott Carey 2009-07-30 19:59:46 Re: PostgreSQL 8.4 performance tuning questions
Previous Message Kevin Grittner 2009-07-30 18:58:46 Re: PostgreSQL 8.4 performance tuning questions