Exploring ‘Load-Average’
February 22nd, 2007 by proj
This is the best exploration of unix ‘load-average‘ that I have ever read. Most commonly in my experience it has been used as a rough metric based on machine quality of load, which is to say, almost completely worthless metric. The basic idea is that it is a running average of the CPU run queue length. At some point I’m going to have to figure out what the state of the art linux kernel/process profiling tools look like. Hopefully they’re better or at parity with windows/perfmon.
Related posts:






