On Thu, 30 Aug 2001, Eiji Kawai wrote:
> I have a question about the minimum hit ratio required at the
> cache-off. I read the meeting minutes of the 3rd cache-off and
> found some discussion about this topic. However, the offered hit
> ratio with webaxe-4 workload is much different from that with
> polymix-3/4 workload. Are there some rules about the minimum hit
> ratio with webaxe-4?
Good question! No minimum hit ratio with WebAxe has been set. I would
suggest to use about the same proportion as PolyMix is using: a box
can "miss" about half of the offered hits. Thus, for WebAxe-4 the
minimum would be about 35%.
If there are no objections, to simplify the rules and automate the
process, we will update the rules to read:
"A test is considered failed if during any measurement phase,
measured hit ratio falls below 50% of the offered hit ratio
and/or if during any measurement phase, measured mean response
time exceeds no-proxy response time by at least 100%"
Thank you for catching this documentation bug!
Alex.
This archive was generated by hypermail 2b29 : Mon Feb 06 2006 - 12:00:19 MST