From: Sylvain Munaut <s.munaut@whatever-company.com>
To: xen-devel@lists.xen.org
Subject: IO speed limited by size of IO request (for RBD driver)
Date: Tue, 23 Apr 2013 15:33:28 +0200 [thread overview]
Message-ID: <CAF6-1L5fePSH1eddGSa5e9ZChuNp3Gdph-wrfjJo75Ts=9Ra4Q@mail.gmail.com> (raw)
Hi,
I was observing a pretty severe performance impact when using Xen VM
with RBD (Ceph) backed storage, especially when doing large sequential
access.
And I think I finally found a major cause for it: even large user
space requests seem to be split into small requests of 11 * 4096
bytes. ( 44k )
This is caused by :
blk_queue_max_segments(rq, BLKIF_MAX_SEGMENTS_PER_REQUEST);
blk_queue_max_segment_size(rq, PAGE_SIZE);
What's the impact of modifying those ?
I've seen some justification for the BLKIF_MAX_SEGMENTS_PER_REQUEST
limit, but why limit segment size to page_size ?
Cheers,
Sylvain
next reply other threads:[~2013-04-23 13:33 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-04-23 13:33 Sylvain Munaut [this message]
2013-04-23 13:41 ` IO speed limited by size of IO request (for RBD driver) Steven Haigh
2013-04-23 14:06 ` Roger Pau Monné
2013-04-23 14:15 ` Sylvain Munaut
2013-04-25 13:00 ` Sylvain Munaut
[not found] ` <51769B9D.4000708@crc.id.au>
[not found] ` <51769CFD.7020907@citrix.com>
[not found] ` <51769E1E.6040902@crc.id.au>
[not found] ` <5176A19A.2010802@citrix.com>
[not found] ` <5176A440.8040303@crc.id.au>
[not found] ` <5176A520.5030503@citrix.com>
[not found] ` <5176A61F.6050607@crc.id.au>
[not found] ` <5176A6DD.5000404@citrix.com>
[not found] ` <5176AFF9.4020003@crc.id.au>
[not found] ` <5176B237.8020803@citrix.com>
[not found] ` <5176C073.3050409@crc.id.au>
[not found] ` <5176CF56.8000505@citrix.com>
[not found] ` <5176DB88.1070200@crc.id.au>
[not found] ` <517A89DA.3030804@citrix.com>
2013-04-26 14:16 ` Steven Haigh
2013-04-27 1:57 ` Steven Haigh
2013-04-27 7:06 ` Roger Pau Monné
2013-04-27 7:51 ` Steven Haigh
2013-04-27 8:35 ` Steven Haigh
2013-04-29 8:38 ` Roger Pau Monné
2013-04-29 19:26 ` Steven Haigh
2013-04-29 19:47 ` Steven Haigh
2013-04-30 10:07 ` Felipe Franciosi
2013-04-30 10:38 ` Steven Haigh
2013-05-08 8:20 ` Steven Haigh
2013-05-08 8:33 ` Roger Pau Monné
2013-05-08 8:47 ` Steven Haigh
2013-05-08 10:32 ` Steven Haigh
2013-05-08 10:45 ` Roger Pau Monné
2013-05-08 11:14 ` Felipe Franciosi
2013-05-22 20:13 ` Konrad Rzeszutek Wilk
2013-05-23 7:22 ` Felipe Franciosi
2013-05-24 14:29 ` Konrad Rzeszutek Wilk
2013-05-08 12:56 ` Steven Haigh
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='CAF6-1L5fePSH1eddGSa5e9ZChuNp3Gdph-wrfjJo75Ts=9Ra4Q@mail.gmail.com' \
--to=s.munaut@whatever-company.com \
--cc=xen-devel@lists.xen.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;
as well as URLs for NNTP newsgroup(s).