From: "Kyle Moffett" <kyle@moffetthome.net>
To: "Linus Torvalds" <torvalds@linux-foundation.org>
Cc: "Brandon Casey" <casey@nrlssc.navy.mil>,
"James Pickens" <jepicken@gmail.com>,
"Bruce Fields" <bfields@fieldses.org>,
"Junio C Hamano" <gitster@pobox.com>,
"Git Mailing List" <git@vger.kernel.org>
Subject: Re: hosting git on a nfs
Date: Fri, 14 Nov 2008 09:31:09 -0500 [thread overview]
Message-ID: <f73f7ab80811140631h44f0c712h5656cd6664d91380@mail.gmail.com> (raw)
In-Reply-To: <alpine.LFD.2.00.0811131707090.3468@nehalem.linux-foundation.org>
On Thu, Nov 13, 2008 at 8:15 PM, Linus Torvalds
<torvalds@linux-foundation.org> wrote:
> - It would be really cool to find some way to automatically notice when
> the tree is hot-cached and we might as well just be linear. I don't
> know exactly what it might be, but one of the nice things is that the
> preloading is _entirely_ optimistic, and we could just stop it in the
> middle if we notice that there's no upside.
Perhaps, rather... notice when the tree is *cold*-cached. At that
point you're already in a pseudo-slowpath and starting several threads
won't be noticeable at all compared to the time savings. You could
probably very easily do that by keeping track of the time as you start
with the normal single-threaded scan of directory entries. If the
entries are all equally slow/fast, there most likely won't be any
benefit at all to spawning extra threads. You might use some cutoff
on number of entries to test before you assume everything is hot and
stop worrying about the time. If there's a lot of variability in the
first few thousand then it means some caches are cold and spawning
extra threads will probably help.
Cheers,
Kyle Moffett
next prev parent reply other threads:[~2008-11-14 14:32 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-11-12 9:29 hosting git on a nfs Thomas Koch
2008-11-12 10:10 ` Julian Phillips
2008-11-12 20:31 ` Brandon Casey
2008-11-12 17:36 ` David Brown
2008-11-12 18:14 ` Linus Torvalds
2008-11-13 18:18 ` J. Bruce Fields
2008-11-13 18:32 ` James Pickens
2008-11-13 20:18 ` Linus Torvalds
2008-11-13 21:05 ` Linus Torvalds
2008-11-13 23:23 ` James Pickens
2008-11-13 23:48 ` Linus Torvalds
2008-11-13 23:23 ` Julian Phillips
2008-11-13 23:42 ` Linus Torvalds
2008-11-14 0:04 ` Julian Phillips
2008-11-14 0:14 ` Brandon Casey
2008-11-14 0:38 ` Linus Torvalds
2008-11-14 0:59 ` Pieter de Bie
2008-11-14 1:15 ` Linus Torvalds
2008-11-14 3:33 ` James Pickens
2008-11-14 5:01 ` Linus Torvalds
2008-11-14 13:01 ` Michael J Gruber
2008-11-14 14:31 ` Kyle Moffett [this message]
2008-11-14 18:32 ` Brandon Casey
2008-11-14 19:23 ` Linus Torvalds
2008-11-14 20:14 ` Junio C Hamano
2008-11-14 23:10 ` Linus Torvalds
2008-11-15 12:08 ` [PATCH] Makefile: introduce NO_PTHREADS Junio C Hamano
2008-11-15 17:15 ` Linus Torvalds
2008-11-17 10:03 ` Mike Ralphson
2008-11-17 10:18 ` Junio C Hamano
2008-11-17 10:34 ` Johannes Sixt
2008-11-17 10:45 ` Mike Ralphson
2008-11-17 11:25 ` Johannes Sixt
2008-12-01 8:29 ` Johannes Sixt
2008-12-01 8:48 ` dhruva
2008-12-01 9:57 ` Mike Ralphson
2008-12-01 16:09 ` Mike Ralphson
2008-12-01 16:13 ` Mike Ralphson
2008-12-02 7:41 ` Johannes Sixt
2008-12-03 2:18 ` Junio C Hamano
2008-11-17 16:38 ` Junio C Hamano
2008-11-17 16:47 ` Linus Torvalds
2008-11-17 17:01 ` Fix index preloading for racy dirty case Linus Torvalds
2008-11-17 16:41 ` [PATCH] Makefile: introduce NO_PTHREADS Junio C Hamano
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=f73f7ab80811140631h44f0c712h5656cd6664d91380@mail.gmail.com \
--to=kyle@moffetthome.net \
--cc=bfields@fieldses.org \
--cc=casey@nrlssc.navy.mil \
--cc=git@vger.kernel.org \
--cc=gitster@pobox.com \
--cc=jepicken@gmail.com \
--cc=torvalds@linux-foundation.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).