From: T Makphaibulchoke <tmac@hp.com>
To: tytso@mit.edu, adilger.kernel@dilger.ca, viro@zeniv.linux.org.uk,
linux-ext4@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-fsdevel@vger.kernel.org
Cc: aswin@hp.com, torvalds@linux-foundation.org,
T Makphaibulchoke <tmac@hp.com>
Subject: [PATCH v2 0/2] ext4: increase mbcache scalability
Date: Thu, 22 Aug 2013 09:54:34 -0600 [thread overview]
Message-ID: <1377186876-57291-1-git-send-email-tmac@hp.com> (raw)
In-Reply-To: <1374108934-50550-1-git-send-email-tmac@hp.com>
The patch consists of two parts. The first part introduces higher degree of
parallelism to the usages of the mb_cache and mb_cache_entries and impacts
all ext filesystems.
The second part of the patch further increases the scalablity of an ext4
filesystem by having each ext4 fielsystem allocate and use its own private
mbcache structure, instead of sharing a single mcache structures across all
ext4 filesystems
Here are some of the benchmark results with the changes.
On a 90 core machine:
Here are the performance improvements in some of the aim7 workloads,
---------------------------
| | % increase |
---------------------------
| alltests | 11.85 |
---------------------------
| custom | 14.42 |
---------------------------
| fserver | 21.36 |
---------------------------
| new_dbase | 5.59 |
---------------------------
| new_fserver | 21.45 |
---------------------------
| shared | 12.84 |
---------------------------
For Swingbench dss workload, with 16 GB database,
-------------------------------------------------------------------------------
| Users | 100 | 200 | 300 | 400 | 500 | 600 | 700 | 800 | 900 |
-------------------------------------------------------------------------------
| % imprvoment | 8.46 | 8.00 | 7.35 | -.313| 1.09 | 0.69 | 0.30 | 2.18 | 5.23 |
-------------------------------------------------------------------------------
| % imprvoment |45.66 |47.62 |34.54 |25.15 |15.29 | 3.38 | -8.7 |-4.98 |-7.86 |
| without using| | | | | | | | | |
| shared memory| | | | | | | | | |
-------------------------------------------------------------------------------
For SPECjbb2013, composite run,
--------------------------------------------
| | max-jOPS | critical-jOPS |
--------------------------------------------
| % improvement | 5.99 | N/A |
--------------------------------------------
On an 80 core machine:
The aim7's results for most of the workloads turn out to the same.
Here are the results of Swingbench dss workload,
-------------------------------------------------------------------------------
| Users | 100 | 200 | 300 | 400 | 500 | 600 | 700 | 800 | 900 |
-------------------------------------------------------------------------------
| % imprvoment |-1.79 | 0.37 | 1.36 | 0.08 | 1.66 | 2.09 | 1.16 | 1.48 | 1.92 |
-------------------------------------------------------------------------------
The changes have been tested with ext4 xfstests to verify that no regression
has been introduced.
Changed in v2:
- New performance data
- New diff summary
T Makphaibulchoke (2):
mbcache: decoupling the locking of local from global data
ext4: each filesystem creates and uses its own mc_cache
fs/ext4/ext4.h | 1 +
fs/ext4/super.c | 24 ++--
fs/ext4/xattr.c | 51 +++++----
fs/ext4/xattr.h | 6 +-
fs/mbcache.c | 293 +++++++++++++++++++++++++++++++++++-------------
include/linux/mbcache.h | 10 +-
6 files changed, 269 insertions(+), 116 deletions(-)
--
1.7.11.3
next prev parent reply other threads:[~2013-08-22 15:54 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-07-18 0:55 [PATCH 0/2] ext4: increase mbcache scalability T Makphaibulchoke
2013-08-22 15:54 ` T Makphaibulchoke [this message]
2013-08-22 15:54 ` [PATCH v2 1/2] mbcache: decoupling the locking of local from global data T Makphaibulchoke
2013-08-22 16:53 ` Linus Torvalds
2013-08-22 15:33 ` Thavatchai Makphaibulchoke
2013-08-22 15:54 ` [PATCH v2 2/2] ext4: each filesystem creates and uses its own mc_cache T Makphaibulchoke
2014-01-24 18:31 ` [PATCH v4 0/3] ext4: increase mbcache scalability T Makphaibulchoke
2014-01-24 18:31 ` [PATCH v4 1/3] fs/mbcache.c change block and index hash chain to hlist_bl_node T Makphaibulchoke
2014-01-24 18:31 ` [PATCH v4 2/3] mbcache: decoupling the locking of local from global data T Makphaibulchoke
2014-01-24 18:31 ` [PATCH v4 3/3] ext4: each filesystem creates and uses its own mc_cache T Makphaibulchoke
2014-01-24 21:38 ` [PATCH v4 0/3] ext4: increase mbcache scalability Andi Kleen
2014-01-25 1:13 ` Thavatchai Makphaibulchoke
2014-01-25 6:09 ` Andreas Dilger
2014-01-27 12:27 ` Thavatchai Makphaibulchoke
2014-02-09 19:46 ` Thavatchai Makphaibulchoke
2014-02-11 19:58 ` Thavatchai Makphaibulchoke
2014-02-13 2:01 ` Andreas Dilger
2013-09-04 16:39 ` [PATCH v3 0/2] " T Makphaibulchoke
2013-09-04 16:39 ` [PATCH v3 1/2] mbcache: decoupling the locking of local from global data T Makphaibulchoke
2013-10-30 13:27 ` Theodore Ts'o
2013-10-30 14:42 ` Theodore Ts'o
2013-10-30 17:32 ` Thavatchai Makphaibulchoke
2013-09-04 16:39 ` [PATCH v3 2/2] ext4: each filesystem creates and uses its own mb_cache T Makphaibulchoke
2013-10-30 14:30 ` Theodore Ts'o
2013-09-04 20:00 ` [PATCH v3 0/2] ext4: increase mbcache scalability Andreas Dilger
2013-09-04 15:33 ` Thavatchai Makphaibulchoke
2013-09-05 15:22 ` Thavatchai Makphaibulchoke
2013-09-05 2:35 ` Theodore Ts'o
2013-09-05 9:49 ` Thavatchai Makphaibulchoke
2013-09-06 5:10 ` Andreas Dilger
2013-09-06 12:23 ` Thavatchai Makphaibulchoke
2013-09-10 20:47 ` Andreas Dilger
2013-09-10 21:02 ` Theodore Ts'o
2013-09-10 17:10 ` Thavatchai Makphaibulchoke
2013-09-11 3:13 ` Eric Sandeen
2013-09-11 11:30 ` Theodore Ts'o
2013-09-11 16:49 ` Eric Sandeen
2013-09-11 19:33 ` Eric Sandeen
2013-09-11 20:32 ` David Lang
2013-09-11 20:48 ` Eric Sandeen
2013-09-11 21:25 ` Theodore Ts'o
2013-09-11 20:36 ` Thavatchai Makphaibulchoke
2013-09-12 3:42 ` Eric Sandeen
2014-02-20 18:37 ` [PATCH V5 0/3] " T Makphaibulchoke
2014-02-20 18:37 ` [PATCH V5 1/3] fs/mbcache.c change block and index hash chain to hlist_bl_node T Makphaibulchoke
2014-02-20 18:37 ` [PATCH V5 2/3] mbcache: decoupling the locking of local from global data T Makphaibulchoke
2014-02-20 18:37 ` [PATCH V5 3/3] ext4: each filesystem creates and uses its own mb_cache T Makphaibulchoke
2014-03-12 16:19 ` [PATCH v5 RESEND 0/3] ext4: increase mbcache scalability T Makphaibulchoke
2014-03-12 16:19 ` [PATCH v5 RESEND 1/3] fs/mbcache.c change block and index hash chain to hlist_bl_node T Makphaibulchoke
2014-03-12 16:19 ` [PATCH v5 RESEND 2/3] mbcache: decoupling the locking of local from global data T Makphaibulchoke
2014-03-12 16:19 ` [PATCH v5 RESEND 3/3] ext4: each filesystem creates and uses its own mb_cache T Makphaibulchoke
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=1377186876-57291-1-git-send-email-tmac@hp.com \
--to=tmac@hp.com \
--cc=adilger.kernel@dilger.ca \
--cc=aswin@hp.com \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=torvalds@linux-foundation.org \
--cc=tytso@mit.edu \
--cc=viro@zeniv.linux.org.uk \
/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;
as well as URLs for NNTP newsgroup(s).