On Wed, Jun 22, 2005 at 01:46:39PM -0700, Frank Mayhar said: > On Wed, 2005-06-22 at 16:05 -0400, Walt Reed wrote: > > (slow disk can cause high load average numbers as you spend > > all your time in I/O Wait.) > > Um, no. At least in traditional Unix (meaning System V and the BSDs), > the "load average" is the average length of the run queue. By > definition, if a process is asleep waiting for I/O to complete (as is > the case for disk), it is _not_ on the run queue, and so doesn't > contribute to load average.
On linux (at least RHEL3) it sure as heck does. I see it VERY frequently on my RHEL3 boxes - especially when dealing with huge postgres DB's (300G). For one application, moving off local disk onto a SAN sped up the I/O and my load averages went from 40 to .2 _______________________________________________ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
