9fans - fans of the OS Plan 9 from Bell Labs
 help / color / mirror / Atom feed
From: erik quanstrom <quanstro@quanstro.net>
To: 9fans@9fans.net
Subject: Re: [9fans] simple venti demo:
Date: Thu, 11 Aug 2011 16:19:31 -0400	[thread overview]
Message-ID: <a61fe9563f8317c1618c9b37af9eed24@ladd.quanstro.net> (raw)
In-Reply-To: <CAP6exYK9pwpRctLeuehEMdmVVW73sC7VOxujRq-Gwu9=gCDQQQ@mail.gmail.c>

> > Pay attention to vtsync?  May be not for your mythical multiTB
> > ramflash but in real life syncing on every write is expensive.
>
> are you sure? On a multicore server, why not have a syncing task and a
> serving task? Since all of the arena is in ram, the synciing task will
> not interfere with the serving task, esp. if sata controller and
> network are on different PCI busses.
>
> I don't think the tradeoffs are obvious at all.

that doesn't sound synchronous to me.  what i think of when
i think of flush on write is that the i/o is done before the reply
to the write.  this has two implications, there's no way to do
any elevatoring, and you take a full round-trip to the disk
delay for each write, no amortization is possible.

i would think that the client is in the best position
to tell the storage when things must be flushed.

it might be best to only write when told to flush and do so in
such a way that it's clear if the transaction has finished.  that way,
if you're really careful and flush caches down to the storage media,
you can recover if things go sideways.

- erik



  parent reply	other threads:[~2011-08-11 20:19 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-08-10 20:53 ron minnich
2011-08-10 21:36 ` Lyndon Nerenberg (VE6BBM/VE7TFX)
2011-08-10 21:47   ` ron minnich
2011-08-10 21:54     ` Lyndon Nerenberg (VE6BBM/VE7TFX)
2011-08-10 21:57       ` erik quanstrom
2011-08-10 22:04       ` Bakul Shah
2011-08-10 22:07         ` Lyndon Nerenberg (VE6BBM/VE7TFX)
2011-08-10 22:12           ` ron minnich
2011-08-11 16:15             ` ron minnich
2011-08-11 16:49               ` David Leimbach
2011-08-11 17:18                 ` Latchesar Ionkov
2011-08-11 19:20               ` Bakul Shah
2011-08-11 19:54               ` Bakul Shah
2011-08-11 20:03                 ` erik quanstrom
2011-08-11 20:07                   ` ron minnich
2011-08-11 20:50                     ` David Leimbach
2011-08-11 21:03                       ` ron minnich
2011-08-11 21:19                       ` Fazlul Shahriar
2011-08-11 21:42                         ` David Leimbach
2011-08-11 21:54                           ` ron minnich
2011-08-11 22:00                             ` David Leimbach
2011-08-11 21:59                           ` Fazlul Shahriar
2011-08-11 20:04                 ` ron minnich
2011-08-11 20:20                   ` Bakul Shah
     [not found]                 ` <CAP6exYK9pwpRctLeuehEMdmVVW73sC7VOxujRq-Gwu9=gCDQQQ@mail.gmail.c>
2011-08-11 20:19                   ` erik quanstrom [this message]
2011-08-12  7:50               ` dexen deVries
2011-08-12 15:45                 ` ron minnich
     [not found]             ` <CAP6exYLmZXLCgBsczjR5e87B095R231WLp3hi0VZOtxgmBd5kw@mail.gmail.c>
2011-08-11 16:49               ` erik quanstrom
2011-08-11 16:55                 ` David Leimbach
2011-08-11 16:56                 ` ron minnich
     [not found]                 ` <CAP6exYJEgJ=KbAtiQR5+qPzYPwWgPU0xtxV1wDY4H_BN_5pWTg@mail.gmail.c>
2011-08-11 17:28                   ` erik quanstrom
2011-08-11 17:37                     ` ron minnich
2011-08-11 18:45                       ` David Leimbach
     [not found]                     ` <CAP6exY+0wGYT2ojbmm+8F1R48w0tycgRTR3KGTm_yki9A7cmSw@mail.gmail.c>
2011-08-11 17:58                       ` erik quanstrom
2011-08-11  3:05           ` David Leimbach
2011-08-11  6:21             ` ron minnich
2011-08-11 13:21               ` Eric Van Hensbergen
2011-08-11 15:09               ` Charles Forsyth
2011-08-11 15:16                 ` Eric Van Hensbergen
2011-08-11 17:06                 ` Salman Aljammaz
2011-08-11 18:33             ` Lyndon Nerenberg (VE6BBM/VE7TFX)
2011-08-11 18:46               ` David Leimbach
2011-08-12 16:18 ` Russ Cox
2011-08-12 18:29   ` Wes Kussmaul

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=a61fe9563f8317c1618c9b37af9eed24@ladd.quanstro.net \
    --to=quanstro@quanstro.net \
    --cc=9fans@9fans.net \
    /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).