lkml.org 
[lkml]   [2000]   [Aug]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Degrading disk read performance under 2.2.16
Mark,

On Sat, 12 Aug 2000, Mark Hahn wrote:
> > > readahead is effecting things (shouldnt, but...)
> >
> > mke2fs only allows block sizes of 1K, 2K or 4K, so I can't make the
> > blocksize any larger...
>
> I meant "size of block being read", rather than filesystem block size.

I've now tested it with 2.2.16 and 8k block reading size - sorry about the
misunderstanding...

Dir Size BlkSz Thr# Read (CPU%) Write (CPU%) Seeks (CPU%)
----- ------ ------- ---- ------------- -------------- --------------
/mnt/ 256 8192 1 26.8839 10.2% 25.6472 22.4% 141.201 1.07%
/mnt/ 256 8192 2 20.4838 8.40% 25.6120 22.8% 136.777 0.80%
/mnt/ 256 8192 4 13.1292 6.09% 25.5942 23.0% 137.310 0.72%
/mnt/ 256 8192 8 10.5354 5.56% 25.4997 23.3% 139.946 0.74%
/mnt/ 256 8192 16 8.48961 5.23% 25.4394 23.4% 142.674 0.81%
/mnt/ 256 8192 32 6.97233 5.51% 25.2732 23.3% 144.932 0.91%

This doesn't appear to make things any better...

> > > so when this is running, does "vmstat 1" indicate that the system's
> > > swapping? I'm guessing so, that it's trying to scavenge pages from
> > > live processes, ineffectually (since they're live), and thrashing.
> >
> > After going to all the trouble of testing this out, I remembered that I
> > had not allocated any swap so that it couldn't influence the results at
> > all! d'oh!
>
> Linux is designed to have swap. I doubt anyone cares about how it
> behaves if you cripple it.

Since this is designed to test raw disk performance, I wanted to reduce
any other factors that might influence it. This includes reducing disk
caching/buffering (by lowering memory) other disk activity (removing
swap). It means I know it's not thrashing pages.

The fact remains that disk performance is much worse under 2.2.16 and
heavy loads than under 2.2.15 - what I was trying to find out was what
was causing it. Turning off swap under 2.2.15 doesn't seem to affect
the performance at all, so /something/ has still gone wrong with the
disk subsystem.

The other reason I've turned off swap is because we don't generally use it
with our RAID boxes - there are still a number or really serious issues
running RAID and swap - and if you run just the ext2 partitions on RAID,
then the machine will still die with a failed disk if it swaps!

> > I think I've located the problem... (with kernel 2.4, anyway) - it is
> > refusing to use DMA.
>
> CONFIG_ option.

Thanks, I have corrected my mistake now. Disk read performance (compared
to 2.2.16) is down for one thread, but decays much less quickly as threads
scale up.

> > hdc: [PTBL] [4982/255/63] hdc1
>
> geometry, initially read from bios and/or disk, is overridden by
> values inferred from the partition table.

Thanks.

> > Anybody have any suggestions about how to get DMA working? Is it a problem
> > with the IDE controller?
>
> you didn't choose exactly the right ide CONFIG_ options.

Yup, that's the cause behind the 2.4 probs.

Regards,

Corin

/------------------------+-------------------------------------\
| Corin Hartland-Swann | Direct: +44 (0) 20 7544 4676 |
| Commerce Internet Ltd | Mobile: +44 (0) 79 5854 0027 |
| 22 Cavendish Buildings | Tel: +44 (0) 20 7491 2000 |
| Gilbert Street | Fax: +44 (0) 20 7491 2010 |
| Mayfair | Web: http://www.commerce.uk.net/ |
| London W1K 5HJ | E-Mail: cdhs@commerce.uk.net |
\------------------------+-------------------------------------/


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:58    [W:0.061 / U:0.068 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site