l i n u x - u s e r s - g r o u p - o f - d a v i s
L U G O D
 
Next Meeting:
April 21: Google Glass
Next Installfest:
TBD
Latest News:
Mar. 18: Google Glass at LUGOD's April meeting
Page last updated:
2005 May 20 07:09

The following is an archive of a post made to our 'vox-tech mailing list' by one of its subscribers.

Report this post as spam:

(Enter your email address)
Re: [vox-tech] Raid 1 - Fresh Install - Drives seem slow
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [vox-tech] Raid 1 - Fresh Install - Drives seem slow



on Fri, May 13, 2005 at 03:35:26PM -0500, Jay Strauss (me@heyjay.com) wrote:
> Hi,
> 
> I just installed Sarge on an old PPro 180Mhz, with / on a RAID1 device 
> and everything else on RAID1/VG/LVGs (which is very easy using the new 
> Debian installer).  I'm building this machine to be an offsite backup 
> machine.

Note that that's about 2.8 MiB/s net rate, if I'm doing my math right.
Depending on what-all else is going on, that's not _too_ unreasonbly
slow, though it's not blazingly fast either.  Disk contention?

 
> It took 10hrs to sync my 100GiB partition, the first time, which seems a 
> bit slow.

Can you define "sync"?  Are you rsyncing one system to another?
Locally/remotely?  Simply running 'sync'?  or what?

If you're doing anything which requires significant CPU usage (say,
compression or encryption), that PPro's going to be relatively pokey.
 
> I'm using 2 Seagate Barracuda 7200.7 ST3160021A 160GB 7200 RPM IDE
> Ultra ATA100 Hard Drive drives, each as the master on separate
> controllers (one of the drives has a CD as a slave)
 
> Googling, I've found a couple of tests like:
> 
> time dd if=/dev/hda of=/dev/null bs=64k count=10000
> 
> which takes like 55 seconds (it only takes roughly 25 seconds on various 
> other machines I have including an old P133)
> 
> I also ran: hdparm -tT /dev/hda, as described in the windows thread by 
> Karsten, and it came in at 28 MB/s

Decent, not great.
 
> - Does the processor speed have anything to do with I/O performance?

Not directly.  Processor speed on blocking tasks (encryption, find/sort,
compression, decompression) can, however.

> - What other tests might I run to prove it's slow?

You've hit most of it.  Clarifying just what you were doing would be
helpful.

I ran across a bad disk (IIRC Maxtor) which was netting ~100 *KiB*/s
last year.  This was sufficiently slow that network operations were
notably lagged.  I confirmed speed issues w/ hdparm and swapped it out
for a faster drive for much better performance (60-80 MiB/s).



Peace.

-- 
Karsten M. Self <kmself@ix.netcom.com>        http://kmself.home.netcom.com/
 What Part of "Gestalt" don't you understand?
    I sincerely believe people talk about me. Mine would be a pretty
    meaningless existance if they didn't.

Attachment: signature.asc
Description: Digital signature

_______________________________________________
vox-tech mailing list
vox-tech@lists.lugod.org
http://lists.lugod.org/mailman/listinfo/vox-tech


LinkedIn
LUGOD Group on LinkedIn
Sign up for LUGOD event announcements
Your email address:
facebook
LUGOD Group on Facebook
'Like' LUGOD on Facebook:

Hosting provided by:
Sunset Systems
Sunset Systems offers preconfigured Linux systems, remote system administration and custom software development.

LUGOD: Linux Users' Group of Davis
PO Box 2082, Davis, CA 95617
Contact Us

LUGOD is a 501(c)7 non-profit organization
based in Davis, California
and serving the Sacramento area.
"Linux" is a trademark of Linus Torvalds.

Sponsored in part by:
Appahost Applications
For a significant contribution towards our projector, and a generous donation to allow us to continue meeting at the Davis Library.