* reconstruction -- where is the upper bound?
@ 2003-10-28 19:55 mjstumpf
2003-10-28 19:58 ` seth vidal
0 siblings, 1 reply; 2+ messages in thread
From: mjstumpf @ 2003-10-28 19:55 UTC (permalink / raw)
To: linux-raid
I'm considering my options; currently I've got a server running on a
p2-400. What causes the upper bound--IO bandwidth or CPU speed--with
regard to reconstruction of an array (say, a failed drive)?
I forget the numbers I usually see, because I do this a lot. But it seems
to me it has more to do with IO bandwidth than CPU horsepower. I've done
this with both SCSI and ATA, but currently I'm referring to an ATA system.
True?
Is there anything I can do to speed this up?
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: reconstruction -- where is the upper bound?
2003-10-28 19:55 reconstruction -- where is the upper bound? mjstumpf
@ 2003-10-28 19:58 ` seth vidal
0 siblings, 0 replies; 2+ messages in thread
From: seth vidal @ 2003-10-28 19:58 UTC (permalink / raw)
To: mjstumpf; +Cc: linux-raid
On Tue, 2003-10-28 at 14:55, mjstumpf wrote:
> I'm considering my options; currently I've got a server running on a
> p2-400. What causes the upper bound--IO bandwidth or CPU speed--with
> regard to reconstruction of an array (say, a failed drive)?
>
> I forget the numbers I usually see, because I do this a lot. But it seems
> to me it has more to do with IO bandwidth than CPU horsepower. I've done
> this with both SCSI and ATA, but currently I'm referring to an ATA system.
>
> True?
>
> Is there anything I can do to speed this up
cd /proc/sys/dev/raid/
cat speed_limit_max
echo numberinKB/s > speed_limit_max
-sv
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2003-10-28 19:58 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-10-28 19:55 reconstruction -- where is the upper bound? mjstumpf
2003-10-28 19:58 ` seth vidal
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).