[ale] usb drives suck

Ed Cashin ecashin at noserose.net
Mon Jan 25 20:59:05 EST 2010


On Mon, Jan 25, 2010 at 2:34 PM, Geoffrey <lists at serioustechnology.com> wrote:
> Ed Cashin wrote:
...
>> So there should be plenty of CPU for other processes while
>> the system waits for the I/O to finish.  I don't know why the
>> load average (running average count of processes that would
>> like to run but have to wait) would go up unless there were lots
>> of processes that couldn't run until the I/O finished.
>
> I don't get that either.  Right now, load average is 7.5, 6.49, 4.6
> I'm running an rsync of /home to an external usb drive.
>
> I guarantee you, once it finishes, load average will drop below 1.

The software arbitration aaron talks about would account for that,
but in that case, I'd expect the system, irq, and/or softirq buckets
to be high and iowait low, reflecting the fact that the CPU isn't
idle but is busy doing the arbitration.

Oh well.  It's a curiosity, but I'm not going to look into it further, so
if anybody finds out more, please let us know.

-- 
  Ed Cashin <ecashin at noserose.net>
  http://noserose.net/e/
  http://www.coraid.com/



More information about the Ale mailing list