http://blog.insidesystems.net/articles/2007/04/11/more-is-better en-us 40 <p>Not always. I did some more tests, and confirmed that this particular workload actually gets slower as you add more than 2 cores.</p> <code> <pre> sysbench --num-threads=${i} --test=oltp --pgsql-user=bench --pgsql-db=bench --db-driver=pgsql --max-time=60 --max-requests=0 --oltp-read-only=on run </pre> </code> <p>Here&#8217;s the results, with the tests run on absolutely identical hardware. (HP <span class="caps">DL360</span>, 2.00 GHz <span class="caps">E5335</span> Xeon, 10K <span class="caps">RPM SAS</span> drives, 10 GB of <span class="caps">RAM</span>)</p> <p><img src="http://blog.insidesystems.net/files/postgres-fbsd62-v2.png"></p> <p>I know 6.2 isn&#8217;t the greatest for <span class="caps">SMP</span>, but I was still surprised to learn that we could get significantly lower overall performance by adding more processors.</p> <p>I tried to test 7.0, but the 7.0 installer didn&#8217;t seem to recognize the iLO virtual keyboard tonight. Maybe tomorrow will go better.</p> Wed, 11 Apr 2007 03:55:00 -0400 urn:uuid:f1fc3bec-3daf-4d4d-a525-6e48cd2829a5 Kevin Way http://blog.insidesystems.net/articles/2007/04/11/more-is-better FreeBSD System Administration PostgreSQL