From: "Kani, Toshimitsu" <toshi.kani-ZPxbGqLxI0U@public.gmane.org>
To: "snitzer-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org"
<snitzer-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Cc: "linux-nvdimm-y27Ovi1pjclAfugRpC6u6w@public.gmane.org"
<linux-nvdimm-y27Ovi1pjclAfugRpC6u6w@public.gmane.org>,
"linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
<linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
"linux-raid-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
<linux-raid-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
"dm-devel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org"
<dm-devel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
"agk-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org"
<agk-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Subject: Re: dm stripe: add DAX support
Date: Tue, 12 Jul 2016 22:22:05 +0000 [thread overview]
Message-ID: <1468362104.8908.43.camel@hpe.com> (raw)
In-Reply-To: <20160624182859.GD13898-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
On Fri, 2016-06-24 at 14:29 -0400, Mike Snitzer wrote:
>
> BTW, if in your testing you could evaluate/quantify any extra overhead
> from DM that'd be useful to share. It could be there are bottlenecks
> that need to be fixed, etc.
Here are some results from fio benchmark. The test is single-threaded and is
bound to one CPU.
DAX LVM IOPS NOTE
---------------------------------------
Y N 790K
Y Y 754K 5% overhead with LVM
N N 567K
N Y 457K 20% overhead with LVM
DAX: Y: mount -o dax,noatime, N: mount -o noatime
LVM: Y: dm-linear on pmem0 device, N: pmem0 device
fio: bs=4k, size=2G, direct=1, rw=randread, numjobs=1
Among the 5% overhead with DAX/LVM, the new DM direct_access interfaces
account for less than 0.5%.
dm_blk_direct_access 0.28%
linear_direct_access 0.17%
The average latency increases slightly from 0.93us to 0.95us. I think most of
the overhead comes from the submit_bio() path, which is used only for
accessing metadata with DAX. I believe this is due to cloning bio for each
request in DM. There is 12% more L2 miss in total.
Without DAX, 20% overhead is observed with LVM. Average latency increases
from 1.39us to 1.82us. Without DAX, bio is cloned for both data and metadata.
Thanks,
-Toshi
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
WARNING: multiple messages have this Message-ID (diff)
From: "Kani, Toshimitsu" <toshi.kani@hpe.com>
To: "snitzer@redhat.com" <snitzer@redhat.com>
Cc: "linux-raid@vger.kernel.org" <linux-raid@vger.kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"dan.j.williams@intel.com" <dan.j.williams@intel.com>,
"dm-devel@redhat.com" <dm-devel@redhat.com>,
"ross.zwisler@linux.intel.com" <ross.zwisler@linux.intel.com>,
"linux-nvdimm@ml01.01.org" <linux-nvdimm@ml01.01.org>,
"agk@redhat.com" <agk@redhat.com>
Subject: Re: dm stripe: add DAX support
Date: Tue, 12 Jul 2016 22:22:05 +0000 [thread overview]
Message-ID: <1468362104.8908.43.camel@hpe.com> (raw)
In-Reply-To: <20160624182859.GD13898@redhat.com>
On Fri, 2016-06-24 at 14:29 -0400, Mike Snitzer wrote:
>
> BTW, if in your testing you could evaluate/quantify any extra overhead
> from DM that'd be useful to share. It could be there are bottlenecks
> that need to be fixed, etc.
Here are some results from fio benchmark. The test is single-threaded and is
bound to one CPU.
DAX LVM IOPS NOTE
---------------------------------------
Y N 790K
Y Y 754K 5% overhead with LVM
N N 567K
N Y 457K 20% overhead with LVM
DAX: Y: mount -o dax,noatime, N: mount -o noatime
LVM: Y: dm-linear on pmem0 device, N: pmem0 device
fio: bs=4k, size=2G, direct=1, rw=randread, numjobs=1
Among the 5% overhead with DAX/LVM, the new DM direct_access interfaces
account for less than 0.5%.
dm_blk_direct_access 0.28%
linear_direct_access 0.17%
The average latency increases slightly from 0.93us to 0.95us. I think most of
the overhead comes from the submit_bio() path, which is used only for
accessing metadata with DAX. I believe this is due to cloning bio for each
request in DM. There is 12% more L2 miss in total.
Without DAX, 20% overhead is observed with LVM. Average latency increases
from 1.39us to 1.82us. Without DAX, bio is cloned for both data and metadata.
Thanks,
-Toshi
next prev parent reply other threads:[~2016-07-12 22:22 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-24 18:23 [PATCH] dm stripe: add DAX support Toshi Kani
2016-06-24 18:23 ` Toshi Kani
2016-06-24 18:23 ` Toshi Kani
2016-06-24 18:29 ` Mike Snitzer
[not found] ` <20160624182859.GD13898-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2016-06-24 18:42 ` Kani, Toshimitsu
2016-06-24 18:42 ` Kani, Toshimitsu
2016-07-12 22:22 ` Kani, Toshimitsu [this message]
2016-07-12 22:22 ` Kani, Toshimitsu
[not found] ` <1468362104.8908.43.camel-ZPxbGqLxI0U@public.gmane.org>
2016-07-13 2:01 ` Mike Snitzer
2016-07-13 2:01 ` Mike Snitzer
[not found] ` <20160713020100.GA5872-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2016-07-13 15:03 ` Kani, Toshimitsu
2016-07-13 15:03 ` Kani, Toshimitsu
[not found] ` <1468422189.8908.53.camel-ZPxbGqLxI0U@public.gmane.org>
2016-07-21 0:01 ` Mike Snitzer
2016-07-21 0:01 ` Mike Snitzer
[not found] ` <20160721000142.GA21913-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2016-07-21 3:03 ` Jens Axboe
2016-07-21 3:03 ` Jens Axboe
2016-06-24 19:58 ` [PATCH] " kbuild test robot
2016-06-24 19:58 ` kbuild test robot
2016-06-24 19:58 ` kbuild test robot
[not found] ` <201606250312.OzHHesq4%fengguang.wu-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2016-06-24 20:02 ` Mike Snitzer
2016-06-24 20:02 ` Mike Snitzer
[not found] ` <20160624200239.GA15587-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2016-06-24 20:46 ` Kani, Toshimitsu
2016-06-24 20:46 ` Kani, Toshimitsu
2016-06-24 19:59 ` [PATCH] " kbuild test robot
2016-06-24 19:59 ` kbuild test robot
2016-06-24 19:59 ` kbuild test robot
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=1468362104.8908.43.camel@hpe.com \
--to=toshi.kani-zpxbgqlxi0u@public.gmane.org \
--cc=agk-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=dm-devel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-nvdimm-y27Ovi1pjclAfugRpC6u6w@public.gmane.org \
--cc=linux-raid-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=snitzer-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.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 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.