From: Ric Wheeler <ricwheeler@gmail.com>
To: Eric Sandeen <sandeen@sandeen.net>,
Dave Chinner <david@fromorbit.com>,
Ilya Dryomov <idryomov@gmail.com>
Cc: xfs <linux-xfs@vger.kernel.org>, Mark Nelson <mnelson@redhat.com>,
Eric Sandeen <sandeen@redhat.com>,
Mike Snitzer <snitzer@redhat.com>
Subject: Re: [PATCH] mkfs.xfs: don't go into multidisk mode if there is only one stripe
Date: Thu, 29 Nov 2018 08:53:39 -0500 [thread overview]
Message-ID: <0290ec9f-ab2b-7c1b-faaf-409d72f99e5f@gmail.com> (raw)
In-Reply-To: <36bc3f17-e7d1-ce8b-2088-36ff5d7b1e8b@sandeen.net>
On 10/6/18 8:14 PM, Eric Sandeen wrote:
> On 10/6/18 6:20 PM, Dave Chinner wrote:
>>> Can you give an example of a use case that would be negatively affected
>>> if this heuristic was switched from "sunit" to "sunit < swidth"?
>> Any time you only know a single alignment characteristic of the
>> underlying multi-disk storage. e.g. hardware RAID0/5/6 that sets
>> iomin = ioopt, multi-level RAID constructs where only the largest
>> alignment requirement is exposed, RAID1 devices exposing their chunk
>> size, remote replication chunk alignment (because remote rep. is
>> slow and so we need more concurrency to keep the pipeline full),
>> etc.
> So the tl;dr here is "given any iomin > 512, we should infer low seek
> latency and parallelism and adjust geometry accordingly?"
>
> -Eric
Chiming in late here, but I do think that every decade or two (no disrespect to
xfs!), it is worth having a second look at how the storage has changed under us.
The workload that has lots of file systems pounding on a shared device for
example is one way to lay out container storage.
No argument about documenting how to fix this with command line tweaks for now,
but maybe this would be a good topic for the next LSF/MM shared track of file &
storage people to debate?
Ric
next prev parent reply other threads:[~2018-11-30 0:59 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-04 17:58 [PATCH] mkfs.xfs: don't go into multidisk mode if there is only one stripe Ilya Dryomov
2018-10-04 18:33 ` Eric Sandeen
2018-10-04 18:56 ` Ilya Dryomov
2018-10-04 22:29 ` Dave Chinner
2018-10-05 11:27 ` Ilya Dryomov
2018-10-05 13:51 ` Eric Sandeen
2018-10-05 23:27 ` Dave Chinner
2018-10-06 12:17 ` Ilya Dryomov
2018-10-06 23:20 ` Dave Chinner
2018-10-07 0:14 ` Eric Sandeen
2018-11-29 13:53 ` Ric Wheeler [this message]
2018-11-29 21:48 ` Dave Chinner
2018-11-29 23:53 ` Ric Wheeler
2018-11-30 2:25 ` Dave Chinner
2018-11-30 18:00 ` block layer API for file system creation - when to use multidisk mode Ric Wheeler
2018-11-30 18:00 ` Ric Wheeler
2018-11-30 18:05 ` Mark Nelson
2018-11-30 18:05 ` Mark Nelson
2018-12-01 4:35 ` Dave Chinner
2018-12-01 4:35 ` Dave Chinner
2018-12-01 20:52 ` Ric Wheeler
2018-12-01 20:52 ` Ric Wheeler
2018-10-07 13:54 ` [PATCH] mkfs.xfs: don't go into multidisk mode if there is only one stripe Ilya Dryomov
2018-10-10 0:28 ` Dave Chinner
2018-10-05 14:50 ` Mike Snitzer
2018-10-05 14:55 ` Eric Sandeen
2018-10-05 17:21 ` Ilya Dryomov
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=0290ec9f-ab2b-7c1b-faaf-409d72f99e5f@gmail.com \
--to=ricwheeler@gmail.com \
--cc=david@fromorbit.com \
--cc=idryomov@gmail.com \
--cc=linux-xfs@vger.kernel.org \
--cc=mnelson@redhat.com \
--cc=sandeen@redhat.com \
--cc=sandeen@sandeen.net \
--cc=snitzer@redhat.com \
/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.