public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: jbarnes@sgi.com (Jesse Barnes)
To: Jens Axboe <axboe@suse.de>
Cc: linux-kernel@vger.kernel.org
Subject: Re: [PATCH] backing dev unplugging
Date: Wed, 10 Mar 2004 13:35:09 -0800	[thread overview]
Message-ID: <20040310213509.GA10888@sgi.com> (raw)
In-Reply-To: <20040310210249.GM15087@suse.de>

On Wed, Mar 10, 2004 at 10:02:50PM +0100, Jens Axboe wrote:
> On Wed, Mar 10 2004, Jesse Barnes wrote:
> > On Wed, Mar 10, 2004 at 09:52:37PM +0100, Jens Axboe wrote:
> > > And what fs, if any? The XFS changes probably need someone knowledgable
> > > about XFS internals to verify them.
> > 
> > Yeah, sorry.  I forgot to post CPU time info too (I'm putting that
> > together now).  The benchmark is doing small, direct I/O requests
> > directly to the block devices (e.g. /dev/sdd1).
> 
> Neat thanks, looking forward to seeing them.

Here's a bit more info.  I can probably do some more benchmarking later
if you need it (though sometimes getting time on that machine can be
hard).

The benchmark creates a thread for each block device.  Each one opens
the device with O_DIRECT and starts doing I/O.

10 qla2200 fc controllers, 64 cpus, 112 disks

-------------------------------------
stock BK tree: ~43945 I/Os per second

  [root@revenue sio]# readprofile -m /root/System.map | sort -nr +2 | head -20
  204038 default_idle                             6376.1875
  700889 snidle                                   1825.2318
  234641 cpu_idle                                 488.8354
   25954 blk_run_queues                            45.0590
    4475 dio_bio_end_io                            11.6536
    6137 scsi_end_request                          11.2812
   18002 scsi_request_fn                            9.5350
     447 ia64_spinlock_contention                   6.9844
   19326 __make_request                             5.3446
    3020 sn_dma_flush                               4.4940
     856 generic_unplug_device                      2.4318
     961 scsi_device_unbusy                         2.3101
     360 current_kernel_time                        1.8750
     547 put_io_context                             1.7094
     932 dio_await_one                              1.6181
     409 blk_queue_bounce                           1.5977
    3484 qla2x00_start_scsi                         1.4140
     147 pid_alive                                  1.1484
     436 as_set_request                             1.1354
     283 kmem_cache_alloc                           1.1055

CPU time from vmstat:
 usr sys id wa
   0  15 20 65
   0  19  0 81
   0  17  0 83
   0  19  0 81
   0  16  0 83
   0  18  0 82
   0  18  0 82
   0  13  0 87
   0  16  0 84
   0  12 26 62

-------------------------------------
w/Jens' patch: ~47149 I/Os per second

  [root@revenue sio]# readprofile -m /root/System.map | sort -nr +2 | head -20
  181993 default_idle                             5687.2812
  624772 snidle                                   1627.0104
  209129 cpu_idle                                 435.6854
    4755 dio_bio_end_io                            12.3828
    6593 scsi_end_request                          12.1195
     435 ia64_spinlock_contention                   6.7969
    2959 sn_dma_flush                               4.4033
    7459 scsi_request_fn                            3.9507
     449 blk_run_backing_dev                        3.5078
    1452 scsi_device_unbusy                         3.4904
     250 kobject_put                                2.6042
     590 put_io_context                             1.8438
     995 dio_await_one                              1.6365
     313 current_kernel_time                        1.6302
     361 kmem_cache_alloc                           1.4102
    3435 qla2x00_start_scsi                         1.3941
    5032 __make_request                             1.3674
     480 as_set_request                             1.2500
     603 scsi_put_command                           1.1085
    6019 schedule                                   1.1064

CPU time from vmstat:
 usr sys id wa
   0   5 60 35
   0   8 28 64
   0   8 20 72
   0   8 19 73
   0   8 17 75
   0   8 15 77
   0   8 15 77
   0   8 12 80
   0   8 11 81
   0   8 11 81

  reply	other threads:[~2004-03-10 21:36 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-03-10 12:45 [PATCH] backing dev unplugging Jens Axboe
2004-03-10 19:55 ` Andrew Morton
2004-03-10 20:03   ` Kenneth Chen
2004-03-10 20:20     ` Jens Axboe
2004-03-10 20:45       ` Jesse Barnes
2004-03-10 20:49         ` Jens Axboe
     [not found]           ` <20040310205237.GK15087@suse.de>
2004-03-10 21:01             ` Jesse Barnes
2004-03-10 21:02               ` Jens Axboe
2004-03-10 21:35                 ` Jesse Barnes [this message]
2004-03-10 23:54                   ` Andrew Morton
2004-03-11  0:03                     ` David Mosberger
2004-03-11  6:30                       ` Jesse Barnes
2004-03-10 20:17   ` Jens Axboe
2004-03-15  5:53   ` Kenneth Chen
2004-03-10 20:14 ` Jeff Garzik
2004-03-10 20:19   ` Jens Axboe
2004-03-10 21:00 ` Andrew Morton
2004-03-10 21:02   ` Jens Axboe
2004-03-10 21:40     ` Miquel van Smoorenburg
2004-03-10 23:05       ` Andrew Morton
2004-03-11  0:05         ` Miquel van Smoorenburg
2004-03-11  0:17           ` Andrew Morton
2004-03-11  6:43         ` Jens Axboe
2004-03-10 22:22 ` Nathan Scott
2004-03-10 23:32   ` Steve Lord
2004-03-11  7:05   ` Jens Axboe
2004-03-11  9:14 ` Joe Thornber
2004-03-11  9:16   ` Jens Axboe
2004-03-11 12:17 ` Christophe Saout
2004-03-11 12:22   ` Jens Axboe
2004-03-11 13:11     ` Christophe Saout

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=20040310213509.GA10888@sgi.com \
    --to=jbarnes@sgi.com \
    --cc=axboe@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox