[ILUG] NFS throughput drops with SCSI activity

[Dave Airlie]

I'd go with the swap network-card suggest also ... eepro Intel card or
later 3c900 series (with latest driver).

Dave.

On Thu, 26 Apr 2001, Kenn Humborg wrote:

>
> Machine:    HP Vectra PPro 200MHz
>             dual-proc board with 1 CPU
> Kernel:     2.2.18 uniprocessor
> Ethernet:   Netgear FA311 100BaseT
>             Netgear-supplied driver
> SCSI:       Adaptec AIC-7880U on-board
>             aic7xxx driver
> DISK:       Seagate ST32171N
>
> I'm setting this machine up to pull disk images from an
> NFS server and write them to the disk.  There is no other
> local disk storage available, so the images are pulled
> across NFS piped to gzip and dumped to the disk.
>
> Reading the image and dumping it to /dev/null gives me
> 2MB/s average across the network.  (In fact, it peaks
> around 4.5MB/s, but is very bursty - don't know why, but
> don't really care, since the CPU can only decompress at
> about 2MB/s average anyway.)
>
> dd-ing from /dev/zero to the SCSI disk gives about 6MB/s,
> which is respectable enough.  Given an approx compression
> ratio of 3:1 (as seen in practice) this matches nicely with
> the network and CPU throughput.
>
> However, while writing to the disk, NFS throughput drops
> to about 0.5MB/s and I get "server timed out" & "server OK"
> messages a lot.  Sometimes NFS practically halts.  (Note
> that this happens even when the NFS and disk commands are
> not in the same pipeline, so it's that the pipes are getting
> full.)
>
> I've tried using irqtune to make the NIC's interrupt higher
> priority than the SCSI adapter's, but it makes no difference.
>
> What could be causing the slow-down and how might I fix it?
>
> Thanks,
> Kenn
>
>
>

-- 
David Airlie, Software Engineer
http://www.skynet.ie/~airlied / airlied at skynet.ie
pam_smb / Linux DecStation / Linux VAX / ILUG person