On Mon, Jun 01, 2009 at 11:07:42AM +0800, Hisashi Hifumi wrote: > > At 12:02 09/06/01, Wu Fengguang wrote: > >On Mon, Jun 01, 2009 at 10:51:56AM +0800, Hisashi Hifumi wrote: > >> > >> At 11:37 09/06/01, Wu Fengguang wrote: > >> >On Wed, May 27, 2009 at 11:06:37AM +0800, Hisashi Hifumi wrote: > >> >> > >> >> At 11:57 09/05/27, Wu Fengguang wrote: > >> >> >On Wed, May 27, 2009 at 10:47:47AM +0800, Hisashi Hifumi wrote: > >> >> >> > >> >> >> At 11:36 09/05/27, Wu Fengguang wrote: > >> >> >> >On Wed, May 27, 2009 at 10:21:53AM +0800, Hisashi Hifumi wrote: > >> >> >> >> > >> >> >> >> At 11:09 09/05/27, Wu Fengguang wrote: > >> >> >> >> >On Wed, May 27, 2009 at 08:25:04AM +0800, Hisashi Hifumi wrote: > >> >> >> >> >> > >> >> >> >> >> At 08:42 09/05/27, Andrew Morton wrote: > >> >> >> >> >> >On Fri, 22 May 2009 10:33:23 +0800 > >> >> >> >> >> >Wu Fengguang <fengguang.wu@xxxxxxxxx> wrote: > >> >> >> >> >> > > >> >> >> >> >> >> > I tested above patch, and I got same performance number. > >> >> >> >> >> >> > I wonder why if (PageUptodate(page)) check is there... > >> >> >> >> >> >> > >> >> >> >> >> >> Thanks! This is an interesting micro timing behavior that > >> >> >> >> >> >> demands some research work. The above check is to confirm > >if it's > >> >> >> >> >> >> the PageUptodate() case that makes the difference. So why > >that case > >> >> >> >> >> >> happens so frequently so as to impact the performance? > >Will it also > >> >> >> >> >> >> happen in NFS? > >> >> >> >> >> >> > >> >> >> >> >> >> The problem is readahead IO pipeline is not running smoothly, > >> >which is > >> >> >> >> >> >> undesirable and not well understood for now. > >> >> >> >> >> > > >> >> >> >> >> >The patch causes a remarkably large performance increase. A 9% > >> >> >> >> >> >reduction in time for a linear read? I'd be surprised if the > >workload > >> >> >> >> >> > >> >> >> >> >> Hi Andrew. > >> >> >> >> >> Yes, I tested this with dd. > >> >> >> >> >> > >> >> >> >> >> >even consumed 9% of a CPU, so where on earth has the kernel > >gone to? > >> >> >> >> >> > > >> >> >> >> >> >Have you been able to reproduce this in your testing? > >> >> >> >> >> > >> >> >> >> >> Yes, this test on my environment is reproducible. > >> >> >> >> > > >> >> >> >> >Hisashi, does your environment have some special configurations? > >> >> >> >> > >> >> >> >> Hi. > >> >> >> >> My testing environment is as follows: > >> >> >> >> Hardware: HP DL580 > >> >> >> >> CPU:Xeon 3.2GHz *4 HT enabled > >> >> >> >> Memory:8GB > >> >> >> >> Storage: Dothill SANNet2 FC (7Disks RAID-0 Array) > >> >> >> > > >> >> >> >This is a big hardware RAID. What's the readahead size? > >> >> >> > > >> >> >> >The numbers look too small for a 7 disk RAID: > >> >> >> > > >> >> >> > > #dd if=testdir/testfile of=/dev/null bs=16384 > >> >> >> > > > >> >> >> > > -2.6.30-rc6 > >> >> >> > > 1048576+0 records in > >> >> >> > > 1048576+0 records out > >> >> >> > > 17179869184 bytes (17 GB) copied, 224.182 seconds, 76.6 MB/s > >> >> >> > > > >> >> >> > > -2.6.30-rc6-patched > >> >> >> > > 1048576+0 records in > >> >> >> > > 1048576+0 records out > >> >> >> > > 17179869184 bytes (17 GB) copied, 206.465 seconds, 83.2 MB/s > >> >> >> > > >> >> >> >I'd suggest you to configure the array properly before coming back to > >> >> >> >measuring the impact of this patch. > >> >> >> > >> >> >> > >> >> >> I created 16GB file to this disk array, and mounted to testdir, dd to > >> >> >this directory. > >> >> > > >> >> >I mean, you should get >300MB/s throughput with 7 disks, and you > >> >> >should seek ways to achieve that before testing out this patch :-) > >> >> > >> >> Throughput number of storage array is very from one product to another. > >> >> On my hardware environment I think this number is valid and > >> >> my patch is effective. > >> > > >> >What's your readahead size? Is it large enough to cover the stripe width? > >> > >> Do you mean strage's readahead size? > > > >What's strage? I mean if your RAID's block device file is /dev/sda, then > > > > blockdev --getra /dev/sda > > > >will tell its readahead size in unit of 512 bytes. > > 256 sectors. That's too small! Try this: blockdev --setra 8192 /dev/sda Thanks, Fengguang -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html