* Roberto C. Sanchez <[EMAIL PROTECTED]> [2006-07-11 18:01:24 -0400]:
> Top is quite reliable. The load average represents how many processes > are ready to run. If everything is trying to access the disk, your CPU > utilization will be low (at least less than 100%) and yet you will have > a high load average since many processes will be stuck waiting for the disk. > Would it improve performance in this case to run a kernel with a different preemption model, or would it make more sense to try different i/o schedulers (anticipatory, deadline, or cfq) ? -- Cheers, Dave -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
