* Marcus Kool <marcus.kool@xxxxxxxxxxxxxxx>: >>> I can confirm that. We switched from 2.6 -> 3.0 with no hassle (at 100 >>> requests/s) >>> >> >> I had Adrian benchmark 3.x recently. With his specific RAM-pathways test. >> >> The cutoff for speed seems to be Squid3 reaching 500-650 req/sec and >> Squid 2.6 going past that into the 800-900 req/sec ranges. At a few >> hundred concurrent requests. > > >> Amos > > Do these limits depend on hardware ? Probably :) Otherwise we'd still use our dinky c64. > if yes, what is the hardware that was used to find these limits ? My box: # cat /proc/meminfo MemTotal: 2076656 kB # cat /proc/cpuinfo | grep model model : 4 model name : Intel(R) Xeon(TM) CPU 3.40GHz model : 4 model name : Intel(R) Xeon(TM) CPU 3.40GHz model : 4 model name : Intel(R) Xeon(TM) CPU 3.40GHz model : 4 model name : Intel(R) Xeon(TM) CPU 3.40GHz # df -h Filesystem Size Used Avail Use% Mounted on /dev/cciss/c0d0p5 13G 9.0G 3.8G 71% / tmpfs 1014M 0 1014M 0% /lib/init/rw tmpfs 1014M 4.0K 1014M 1% /dev/shm /dev/cciss/c0d0p8 17G 15G 1.2G 93% /squid-cache0 /dev/cciss/c0d0p9 17G 15G 1.3G 93% /squid-cache1 /dev/cciss/c0d0p10 18G 12G 4.6G 73% /squid-data -- Ralf Hildebrandt (i.A. des IT-Zentrums) Ralf.Hildebrandt@xxxxxxxxxx Charite - Universitätsmedizin Berlin Tel. +49 (0)30-450 570-155 Gemeinsame Einrichtung von FU- und HU-Berlin Fax. +49 (0)30-450 570-962 IT-Zentrum Standort CBF send no mail to plonk@xxxxxxxxxx
