dbaldoni@iinet.net.au wrote:
> Apparently, a general rule of thumb is to use the (Page Faults with
> physical I/O) / (Number of incoming HTTP requests received) ratio; both
Best measured as a delta
Increase in page faults / increase in HTTP requests
taken during peak usage, or measured during the day each X minutes and
plotted...
> Any thoughts on what I should be looking at to provide the system owners
> with an accurate performance measure?
The above, combined with latency information, all plotted in nice
graphs. See MRTG and/or rrdtool for nice graph generators..
-- Henrik Nordstrom Squid HackerReceived on Fri Jun 15 2001 - 05:16:43 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:00:45 MST