git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Drew Northup <n1xim.email@gmail.com>
To: Thomas Koch <thomas@koch.ro>
Cc: Git Mailing List <git@vger.kernel.org>
Subject: Re: git fuse
Date: Wed, 25 Dec 2013 18:01:12 -0500	[thread overview]
Message-ID: <CAM9Z-nmVaBDcRABMRJGLdFVDSLRGL31jMS9KpaWiQ4y_SYONdg@mail.gmail.com> (raw)
In-Reply-To: <201312192215.01103.thomas@koch.ro>

On Thu, Dec 19, 2013 at 4:15 PM, Thomas Koch <thomas@koch.ro> wrote:
> Hi,
>
> I'm thinking about using Git for web application deployment and thought that I
> wouldn't even need to checkout a worktree if I could access the bare git repo
> via fuse.

Thomas,
Given that gitweb (and more than a few other similar tools) doesn't
use FUSE and works on bare repositories I question why you think you'd
need (or want) to introduce the additional layer.

> What would be the performance impact? Once the files are in the filesystem cache
> it shouldn't matter anymore, how fast the git fuse layer is, should it?

If your concern is caching, it should be implemented upon already
rendered / prepared objects whenever possible--frequently not a file
system level implementation. This is a design concern that has no
specific requirement whatsoever to do with FUSE (or for that matter,
with most of the VFS layer when accessing raw disk--as VFS itself is
highly optimized already).

I am not qualified to answer questions about FUSE use of the VFS cache.

I would strongly consider using a tool such as Varnish if you are
concerned about performance when serving static or semi-static content
(may it reside in a Git repo or elsewhere). The architectural concerns
of your web-app should probably guide your design in this sort of
direction anyway if it will be expected to sustain notable throughput
over time.

I hope that helps.

(If you want to discuss this further it would be worth considering if
your questions are Git questions or web-app design questions. The
latter should be taken elsewhere.)

-- 
-Drew Northup
--------------------------------------------------------------
"As opposed to vegetable or mineral error?"
-John Pescatore, SANS NewsBites Vol. 12 Num. 59

      reply	other threads:[~2013-12-25 23:08 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-12-19 21:15 git fuse Thomas Koch
2013-12-25 23:01 ` Drew Northup [this message]

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=CAM9Z-nmVaBDcRABMRJGLdFVDSLRGL31jMS9KpaWiQ4y_SYONdg@mail.gmail.com \
    --to=n1xim.email@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=thomas@koch.ro \
    /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).