All of lore.kernel.org
 help / color / mirror / Atom feed
From: Dave Hansen <haveblue@us.ibm.com>
To: Rik van Riel <riel@conectiva.com.br>
Cc: Badari Pulavarty <pbadari@us.ibm.com>,
	linux-kernel@vger.kernel.org, akpm@zip.com.au,
	Gerrit Huizenga <gerrit@us.ibm.com>,
	Hans-J Tannenberger <hjt@us.ibm.com>,
	Janet Morgan <janetmor@us.ibm.com>,
	Mike Anderson <andmike@us.ibm.com>,
	Martin Bligh <mjbligh@us.ibm.com>
Subject: Re: 2.5.32 IO performance issues
Date: Thu, 29 Aug 2002 12:02:41 -0700	[thread overview]
Message-ID: <3D6E6FD1.8040204@us.ibm.com> (raw)
In-Reply-To: Pine.LNX.4.44L.0208291538470.1857-100000@imladris.surriel.com

Rik van Riel wrote:
> On Thu, 29 Aug 2002, Badari Pulavarty wrote:
>>I am having severe IO performance problems with 2.5.32 (2.5.31 works fine).
>>I was wondering what caused this.
>>
>>As you can see, IO rate went from
>>
>>		384MB/sec with 6% CPU utilization on 2.5.31
>>			to
>>		120MB/sec with 19% CPU utilization on 2.5.32
>>
>>Any idea ?
> 
> 384 MB/s is suspiciously fast.  What kind of disk subsystem
> do you have to achieve that speed ?

Oh, that's nothing! :)

> Hardware: 8x 700MHz P-III, 4 Qlogic FC controllers, 40 disks
> Test: 40 dds on 40 raw devices (40 disks).

One tray of 10 10k RPM disks, per controller.  The controllers are also 
spread across 2 completely separate 64bit/66Mhz PCI busses (not bridged).

   Bus 10, device   8, function  0:
     SCSI storage controller: QLogic Corp. QLA2200 (#2) (rev 5).
       IRQ 23.
       Master Capable.  Latency=96.  Min Gnt=64.
       I/O at 0xb000 [0xb0ff].
       Non-prefetchable 32 bit memory at 0xfe1ff000 [0xfe1fffff].
   Bus 10, device   9, function  0:
     SCSI storage controller: QLogic Corp. QLA2200 (#3) (rev 5).
       IRQ 24.
       Master Capable.  Latency=96.  Min Gnt=64.
       I/O at 0xb100 [0xb1ff].
       Non-prefetchable 32 bit memory at 0xfe1fe000 [0xfe1fefff].
   Bus 13, device   6, function  0:
     SCSI storage controller: QLogic Corp. QLA2200 (#4) (rev 5).
       IRQ 21.
       Master Capable.  Latency=96.  Min Gnt=64.
       I/O at 0xd000 [0xd0ff].
       Non-prefetchable 32 bit memory at 0xfebff000 [0xfebfffff].
   Bus 13, device   7, function  0:
     SCSI storage controller: QLogic Corp. QLA2200 (#5) (rev 5).
       IRQ 22.
       Master Capable.  Latency=96.  Min Gnt=64.
       I/O at 0xd100 [0xd1ff].
       Non-prefetchable 32 bit memory at 0xfebfe000 [0xfebfefff].


-- 
Dave Hansen
haveblue@us.ibm.com


  reply	other threads:[~2002-08-29 19:00 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-08-29 18:20 2.5.32 IO performance issues Badari Pulavarty
2002-08-29 18:39 ` Rik van Riel
2002-08-29 19:02   ` Dave Hansen [this message]
2002-08-29 18:43 ` Andrew Morton
2002-08-29 20:55   ` Badari Pulavarty
2002-08-29 21:53     ` Patrick Mansfield
2002-08-29 22:23       ` Andrew Morton
2002-08-31  9:14         ` Jens Axboe
2002-08-31  9:14         ` Jens Axboe

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=3D6E6FD1.8040204@us.ibm.com \
    --to=haveblue@us.ibm.com \
    --cc=akpm@zip.com.au \
    --cc=andmike@us.ibm.com \
    --cc=gerrit@us.ibm.com \
    --cc=hjt@us.ibm.com \
    --cc=janetmor@us.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mjbligh@us.ibm.com \
    --cc=pbadari@us.ibm.com \
    --cc=riel@conectiva.com.br \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.