Now that I got SQL Bayes storage running with MySQL, I thought I'd do a quick and dirty performance comparison on sa-learn.

I restored backups of 100,512 spam, and 61,653 ham from my current bayes database, then tried "sa-learn --spam --mbox" of 72 recent spams from my spam folder.

Here are the user CPU times for each:
        SQL     6.578s
        DBM     32.798s

I'm running bare mysql, no innodb or anything. I suspect that would increase the INSERT time, but maybe it would speed up the SELECT time. <shrug>

Mojo
--
Morris Jones
Monrovia, CA
http://www.whiteoaks.com
Old Town Astronomers: http://www.otastro.org

Reply via email to