On Tue, Dec 19, 2006 at 11:12:54AM +0000, G.W. Haywood wrote: > > I tried clamav 0.90rc2 and the scantime of clamav minimized to 30-50 > > seconds. [..] > Has anyone else observed such large improvements?
Well, I can only observe the fact that clamav is about 10 times slower than our commercial scanners. These are the statistics for one day begin december somewhere (that I happened to have handy right now) : scanner : avg +- stddev (number of times called) Virus:FPROTD: 0.078 +- 0.447 (7085536 samples) Virus:SOPHIE: 0.059 +- 0.183 (7086708 samples) Virus:CLAMD: 0.787 +- 3.210 (7086846 samples) Compare this to, eg, spamassassin: SpamAssassin: 1.891 +- 1.933 (3194175 samples) And then consider that spamassassin usually only takes 25% cpu, where clam takes nearly 100% cpu. Clamav currently eats more CPU than spamassassin, so currently, adding tests to clamav that could also be done by spamassassin is not to my benefit, CPU-wise... I was hoping that this would change with 0.90, but I haven't tried it on our production platforms yet. -- Jan-Pieter Cornet <[EMAIL PROTECTED]> !! Disclamer: The addressee of this email is not the intended recipient. !! !! This is only a test of the echelon and data retention systems. Please !! !! archive this message indefinitely to allow verification of the logs. !! _______________________________________________ Help us build a comprehensive ClamAV guide: visit http://wiki.clamav.net http://lurker.clamav.net/list/clamav-users.html
