Re: Squid Cache Partition Performance Comparison of Reiserfs and Ext2

Warren,

        Thanks for posting your results! Just a few short non-Squid
specific comments for you or anybody trying to replicate this.

    - PolyMix-1 is relatively old; if nothing else, you may want
      to at least increase the test duration to 4 hours or so
    - always look at the traces to see if the aggregate stats are
      good representation of what was happening (`lx --trace' or
      `ltrace' in the latest release)
    - results with response times above 6 seconds should probably
      ignored or treated with extreme caution (you've got 18sec)
    - results with more that 1% of errors should probably be ignored
      or treated with extreme caution (you've got 5-10%)

In a nutshell, it looks like you severely overloaded your Squid box (or
something else in your setup), and I would not recommend relying on
these results. You need to decrease request rate until you get virtually
no errors (less than 1%) and reasonable response times (at most 2sec for
hits and 3.5sec for misses).

HTH,

Alex.

On Sun, 26 Dec 1999, Warren Togami wrote:

> Squid Cache Partition Performance Comparison of Reiserfs and Ext2
> Workload Simulated by Polygraph 2.2.7 polymix-1.pg
>
> -- ReiserFS Results --
> Full Logs - http://www.togami.com/laven/archive/1999/reiser-ext2/reiser.log.gz
> Full lx Output - http://www.togami.com/laven/archive/1999/reiser-ext2/reiser.lx
>
> req_rate: 47.83
> rep_rate: 47.80
> rptm_mean:hit: 18219.75
> rptm_mean:miss: 19488.74
> err_pcnt: 5.11
>
> -- ext2fs Results --
> Full Logs - http://www.togami.com/laven/archive/1999/reiser-ext2/ext2.log.gz
> Full lx Output - http://www.togami.com/laven/archive/1999/reiser-ext2/ext2.lx
>
> req_rate: 45.51
> rep_rate: 45.44
> rptm_mean:hit: 13994.46
> rptm_mean:miss: 16725.03
> err_pcnt: 9.66
Received on Mon Dec 27 1999 - 10:12:37 MST