Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I would like to benchmark a powerfull webcluster with autobench. The problem is that autobench (which uses httperf) can only use one core, thus limiting the amount of load that a client can put on the web cluster. One core is definatly not enough to saturate a cluster of multiple servers.

autobench ships with autobenchd, which can be run on many clients, to make a "client cluster". My problem is I don't have many clients, but only one powerfull multicore client. I tried to run multiple instances of autobenchd on my client, but then autobench_admin only connects to one of these instances.

I could of course use virtualisation to make multiple virtual clients on my physcial server, but that would waste performance and also take more time to set up. Is there a way to make autobench take advantage of a multicore machine?

share|improve this question
It turns out that tsung ( ) is a much better tools for that purpose. I am using tsung now and dropped autobench. – Isaac Aug 31 '12 at 14:18

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.