All of lore.kernel.org
 help / color / mirror / Atom feed
From: Howard Chu <hyc@symas.com>
To: Gregory Farnum <gfarnum@redhat.com>
Cc: Ric Wheeler <rwheeler@redhat.com>,
	ceph-devel <ceph-devel@vger.kernel.org>
Subject: Re: newstore direction
Date: Fri, 23 Oct 2015 22:23:30 +0100	[thread overview]
Message-ID: <562AA552.702@symas.com> (raw)
In-Reply-To: <CAJ4mKGZ_Xg4hNrm-zjBCJwe4vV1CbALT+C_Ee7cmcpRf=VMKKA@mail.gmail.com>

Gregory Farnum wrote:
> On Fri, Oct 23, 2015 at 7:59 AM, Howard Chu <hyc@symas.com> wrote:
>> If the stream of writes is large enough, you could omit fsync because
>> everything is being forced out of the cache to disk anyway. In that
>> scenario, the only thing that matters is that the writes get forced out in
>> the order you intended, so that an interruption or crash leaves you in a
>> known (or knowable) state vs unknown.
>
> The RADOS storage semantics actually require that we know it's durable
> on disk as well, unfortunately. But ordered writes would probably let
> us batch up commit points in ways that are a lot friendlier for the
> drives!

Ah, that's too bad. LMDB does fine with only ordering, but never mind.

-- 
   -- Howard Chu
   CTO, Symas Corp.           http://www.symas.com
   Director, Highland Sun     http://highlandsun.com/hyc/
   Chief Architect, OpenLDAP  http://www.openldap.org/project/

  reply	other threads:[~2015-10-23 21:23 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-10-19 19:49 newstore direction Sage Weil
2015-10-19 20:22 ` Robert LeBlanc
2015-10-19 20:30 ` Somnath Roy
2015-10-19 20:54   ` Sage Weil
2015-10-19 22:21     ` James (Fei) Liu-SSI
2015-10-20  2:24       ` Chen, Xiaoxi
2015-10-20 12:30         ` Sage Weil
2015-10-20 13:19           ` Mark Nelson
2015-10-20 17:04             ` kernel neophyte
2015-10-21 10:06             ` Allen Samuels
2015-10-21 13:35               ` Mark Nelson
2015-10-21 16:10                 ` Chen, Xiaoxi
2015-10-22  1:09                   ` Allen Samuels
2015-10-20  2:32       ` Varada Kari
2015-10-20  2:40         ` Chen, Xiaoxi
2015-10-20 12:34       ` Sage Weil
2015-10-20 20:18         ` Martin Millnert
2015-10-20 20:32         ` James (Fei) Liu-SSI
2015-10-20 20:39           ` James (Fei) Liu-SSI
2015-10-20 21:20           ` Sage Weil
2015-10-19 21:18 ` Wido den Hollander
2015-10-19 22:40 ` Varada Kari
2015-10-20  0:48 ` John Spray
2015-10-20 20:00   ` Sage Weil
2015-10-20 20:36     ` Gregory Farnum
2015-10-20 21:47       ` Sage Weil
2015-10-20 22:23         ` Ric Wheeler
2015-10-21 13:32           ` Sage Weil
2015-10-21 13:50             ` Ric Wheeler
2015-10-23  6:21               ` Howard Chu
2015-10-23 11:06                 ` Ric Wheeler
2015-10-23 11:47                   ` Ric Wheeler
2015-10-23 14:59                     ` Howard Chu
2015-10-23 16:37                       ` Ric Wheeler
2015-10-23 18:59                       ` Gregory Farnum
2015-10-23 21:23                         ` Howard Chu [this message]
2015-10-20 20:42     ` Matt Benjamin
2015-10-22 12:32     ` Milosz Tanski
2015-10-23  3:16       ` Howard Chu
2015-10-23 13:27         ` Milosz Tanski
2015-10-20  2:08 ` Haomai Wang
2015-10-20 12:25   ` Sage Weil
2015-10-20  7:06 ` Dałek, Piotr
2015-10-20 18:31 ` Ric Wheeler
2015-10-20 19:44   ` Sage Weil
2015-10-20 21:43     ` Ric Wheeler
2015-10-20 19:44   ` Yehuda Sadeh-Weinraub
2015-10-21  8:22   ` Orit Wasserman
2015-10-21 11:18     ` Ric Wheeler
2015-10-21 17:30       ` Sage Weil
2015-10-22  8:31         ` Christoph Hellwig
2015-10-22 12:50       ` Sage Weil
2015-10-22 17:42         ` James (Fei) Liu-SSI
2015-10-22 23:42           ` Samuel Just
2015-10-23  0:10             ` Samuel Just
2015-10-23  1:26             ` Allen Samuels
2015-10-23  2:06         ` Ric Wheeler
2015-10-21 10:06   ` Allen Samuels
2015-10-21 11:24     ` Ric Wheeler
2015-10-21 14:14       ` Mark Nelson
2015-10-21 15:51         ` Ric Wheeler
2015-10-21 19:37           ` Mark Nelson
2015-10-21 21:20             ` Martin Millnert
2015-10-22  2:12               ` Allen Samuels
2015-10-22  8:51                 ` Orit Wasserman
2015-10-22  0:53       ` Allen Samuels
2015-10-22  1:16         ` Ric Wheeler
2015-10-22  1:22           ` Allen Samuels
2015-10-23  2:10             ` Ric Wheeler
2015-10-21 13:44     ` Mark Nelson
2015-10-22  1:39       ` Allen Samuels

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=562AA552.702@symas.com \
    --to=hyc@symas.com \
    --cc=ceph-devel@vger.kernel.org \
    --cc=gfarnum@redhat.com \
    --cc=rwheeler@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.