On Wed, Nov 06, 2002 at 12:28:13PM -0600, steve rader wrote:
> 
>  > From: Net Llama!
>  > OK, thanks for the explanation.  My solaris knowledge is admittedly weak.
>  > Anyway, doesn't iostat do this under Linux?
> 
> Looks like linux iostat still can't report % io wait
> cuz the kernel disk subsystem doesn't count "jiffies"
> (ms or cycles or clock ticks or somesuch?) where
> the cpu is idle and disk requests are outstanding.
> Refer to man iostat and see also /proc/partitions.
> 
> I guess that await from (sysstat) iostat -x might
> sorta kinda help one ferret out disk io bottlenecks?
> But you'd need to do some baselining first?

Good ole sar and friends have been ported to Linux. Sebastian Godard
did the work and bundled it up with a grundle of other stuff in 
sysstat:

http://perso.wanadoo.fr/sebastien.godard/

I don't think it is quite the same as the sar we know and love
from the various *nixen, but perhaps it will prove useful. I know
there are some kernel patches floating about that improve the
quality of performance monitoring.

Kurt
-- 
"We are on the verge: Today our program proved Fermat's next-to-last
theorem."
                -- Epigrams in Programming, ACM SIGPLAN Sept. 1982
_______________________________________________
Linux-users mailing list
[EMAIL PROTECTED]
Unsubscribe/Suspend/Etc -> http://www.linux-sxs.org/mailman/listinfo/linux-users

Reply via email to