I am converting an older system Apache 2.2 using mpm_prefork to Apache 2.4. I think we would have better performance with mpm_worker or possibly mpm_event. My problem is trying to figure out what is the best ratio of threads per process, the correct way to set that value, and the best value for the total number of threads which I think is set by MaxRequestWorkers. I am tuning my cluster of systems for a sudden heavy load of page requests but not a lot of CGI or other complex operations. These are dual processor VM systems. Any suggestions or pointers?
Darryl Baker Sr. System Administrator Distributed Application Platform Services Project Management Office and Enabling Technologies Northwestern | Information Technology 1800 Sherman Ave., Suite 600 Evanston, IL 60201-3715 (847) 467-6674 www.it.northwestern.edu
