On 5/21/07, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
> Be careful with %w... it's not that accurate. If you upgrade your
> e20k to Solaris 10, you'll lose that as iowait is no longer
> calculated (although the %w column is still there for output
> compatibility reasons. %b (% busy) is what yo
> Be careful with %w... it's not that accurate. If you upgrade your
> e20k to Solaris 10, you'll lose that as iowait is no longer
> calculated (although the %w column is still there for output
> compatibility reasons. %b (% busy) is what you should be looking at
> instead.
That's not entir
Your workload has a number of concurrent threads that run batch. Some
number of them can block on I/O at the same time, so you will see the
b column. This will turn up in iostat in the queue length counter
column "actv". So I would expect some correlation between actv and b.
Its not a problem as
On May 18, 2007, at 3:26 PM, Amir Hameed wrote:
I guess what I am looking for is that if the wait, asvc_t, wsvc_t
and %w columns from the iostat lok good then should one pay any
attention to the "b" column from vmstat.
Be careful with %w... it's not that accurate. If you upgrade your
e20k
I guess what I am looking for is that if the wait, asvc_t, wsvc_t and %w
columns from the iostat lok good then should one pay any attention to the "b"
column from vmstat.
This message posted from opensolaris.org
___
perf-discuss mailing list
perf-di
