Nick,

One other thing about this. It's not the scan time that is the most important, it's the CPU utilization. As it turns out, they are roughly related, and clearly clamdscan and F-Prot are the best performers by a mile. Back when I did my tests of these things, those two ran neck and neck, hardly registering any load, and then with a big jump, McAfee and AVG were third and fourth best.

I have minimized the affects by doing AV scanning after the JunkMail scanning, which is behind an Alligate gateway, so many fewer messages get scanned for viruses, but still AV scanning is a major contributer to overall load.

Matt



Nick Hayer wrote:


David Barker wrote:

 Declude has
this process and functionality built in to the Declude product which enables
customers to use a command line scanners like ClamWin

<snip>
Brian Burns port called ClamAV supports clamd which is wicked fast [and free]. This is a different product from ClamWin.

I had done some comparison on the scanners - and had sent this to the Declude Virus list: "I just switched to 4x and noticed in the logs that scan times are recorded -
here are some sample scan times against the same email -
2062ms Clamscan
468ms Mcafee scan.exe
171ms fprot

These relative scan time proportional differences appear to remain the same against other emails.

Switching from clamscan.exe to clamdscan.exe ClamAV averages 15ms against all emails it sees. That is like a factor of 10 faster than fprot its closest performance competitor. Since its free and w/Sanesecurity phish sigs I give it an editors choice :)

It would be nice to see [feature request?] the ms response time for AVG - "

-Nick




---
This E-mail came from the Declude.JunkMail mailing list.  To
unsubscribe, just send an E-mail to [EMAIL PROTECTED], and
type "unsubscribe Declude.JunkMail".  The archives can be found
at http://www.mail-archive.com.





---
This E-mail came from the Declude.JunkMail mailing list.  To
unsubscribe, just send an E-mail to [EMAIL PROTECTED], and
type "unsubscribe Declude.JunkMail".  The archives can be found
at http://www.mail-archive.com.

Reply via email to