On Mon, 10 Jul 2000 Craig_Lowery@Dell.com wrote:
> I'm also experiencing different cache hit ratio results than I did with
> polymix2 (polygraph 2.2.9). I haven't tried altering the reporting tools.
I am assuming that Vincent is using exclusively Polygraph 2.2.9 and
Polymix-2 and observers discrepancy between lx and make_report output.
Lower hit ratio for Polymix-3 compared to PolyMix-2 may be due to the
working set size being larger in PolyMix-3 (you may have seen prior
postings (with workarounds) about that). We will remind about this issue
and offer some tips when 2.5.2 is released.
Alex.
> -----Original Message-----
> From: Vincent Y H Man [mailto:8vyhm@qlink.queensu.ca]
> Sent: Monday, July 10, 2000 10:17 AM
> To: polygraph@ircache.net
> Subject: hit ratio
>
>
> I've been producing reports with polygraph 2.2.9.
>
> Everything seems to be normal except the hit ratio, which is around 38%
> (offered AND measured) This seemed odd to me since polymix-2.pg has a 55%
> recurrence and 80% cachebility ratio.
>
> Then I generated a report using:
>
> /src/lx
>
> and I got a hit ratio of 55% (measured AND offered).
> Anyone have any experience like this?
> Forgive me if I am missing something blatently obvious.
>
> Thanks,
> Vincent
> JKLMicro
>
>
This archive was generated by hypermail 2b29 : Tue Jul 10 2001 - 12:00:14 MDT