git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: Brandon Casey <casey@nrlssc.navy.mil>
To: Nicolas Pitre <nico@cam.org>
Cc: Johannes Schindelin <Johannes.Schindelin@gmx.de>,
	Jan Holesovsky <kendy@suse.cz>, Jakub Narebski <jnareb@gmail.com>,
	git@vger.kernel.org, Junio C Hamano <gitster@pobox.com>
Subject: Re: [PATCH] RFC: git lazy clone proof-of-concept
Date: Thu, 14 Feb 2008 13:41:31 -0600	[thread overview]
Message-ID: <47B4996B.4000900@nrlssc.navy.mil> (raw)
In-Reply-To: <alpine.LFD.1.00.0802101445430.2732@xanadu.home>

Nicolas Pitre wrote:
> On Sun, 10 Feb 2008, Johannes Schindelin wrote:
> 
>> I tried that:
>>
>> $ git config pack.deltaCacheLimit 1
>> $ git config pack.deltaCacheSize 1
>> $ git config pack.windowMemory 2g
> 
> This has nothing to do with repacking memory usage, but even tighter 
> packs can be obtained with:
> 
> 	git config repack.usedeltabaseoffset true
> 
> This is not the default yet.

I have successfully repacked this repo a few times on a 2.1GHz system with 16G.

The smallest attained pack was about 1.45G (1556569742B).

This run took about 7 hours 26 min.

I ran: git repack -a -d -f --window=250 --depth=250

Here are the relevent config entries:
[pack]
        threads = 1
        compression = 9
[repack]
        usedeltabaseoffset = true


Other runs:


* Same as above, but with default compression:

	pack size: 1560624388
	time: 7 hours 11 min

	Not much difference in time or size.


* Multi threaded (250m window)
[pack]
        threads = 4
        windowmemory = 250m
        compression = 9
[repack]
        usedeltabaseoffset = true

	pack size: 1767405703
	time: 3 hours

	First >99% took 50min. Last 10000 objects took 2hours.

* Multi threaded (500m window)
[pack]
        threads = 4
        windowmemory = 500m
        compression = 9
[repack]
        usedeltabaseoffset = true

	pack size: 1640820903
	time: forgot to time, but between 3-4 hours based on file time

	I just received Dscho's email, this is interesting to compare
	with his single threaded result of 1638490531. I wonder if he
	used deltabaseoffset? I think his machine is a little faster
	than this one. So using 4 threads finished twice as fast and
	produced a similar pack size. Actually, the difference could
	just be the compression setting.

* Deeper (git repack -a -d -f --window=250 --depth=500)
[pack]
        threads = 1
        compression = 9
[repack]
        usedeltabaseoffset = true

	pack size: 1578263745
	time: 7 hours 58 min

	Larger pack compared to --depth=250.

-brandon

  reply	other threads:[~2008-02-14 19:43 UTC|newest]

Thread overview: 85+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-02-08 17:28 [PATCH] RFC: git lazy clone proof-of-concept Jan Holesovsky
2008-02-08 18:03 ` Nicolas Pitre
2008-02-09 14:25   ` Jan Holesovsky
2008-02-09 22:05     ` Mike Hommey
2008-02-09 23:38       ` Nicolas Pitre
2008-02-10  7:23     ` Marco Costalba
2008-02-10 12:08       ` Johannes Schindelin
2008-02-10 16:46         ` David Symonds
2008-02-10 17:45           ` Johannes Schindelin
2008-02-10 19:45             ` Nicolas Pitre
2008-02-10 20:32               ` Johannes Schindelin
2008-02-08 18:14 ` Harvey Harrison
2008-02-09 14:27   ` Jan Holesovsky
2008-02-08 18:20 ` Johannes Schindelin
2008-02-08 18:49 ` Mike Hommey
2008-02-08 19:04   ` Johannes Schindelin
2008-02-09 15:06   ` Jan Holesovsky
2008-02-08 19:00 ` Jakub Narebski
2008-02-08 19:26   ` Jon Smirl
2008-02-08 20:09     ` Nicolas Pitre
2008-02-11 10:13       ` Andreas Ericsson
2008-02-12  2:55         ` [PATCH 1/2] pack-objects: Allow setting the #threads equal to #cpus automatically Brandon Casey
2008-02-12  5:53           ` Andreas Ericsson
     [not found]         ` <1202784078-23700-1-git-send-email-casey@nrlssc.navy.mil>
2008-02-12  2:59           ` [PATCH 2/2] pack-objects: Default to zero threads, meaning auto-assign to #cpus Brandon Casey
2008-02-12  4:57             ` Nicolas Pitre
2008-02-08 20:19     ` [PATCH] RFC: git lazy clone proof-of-concept Harvey Harrison
2008-02-08 20:24       ` Jon Smirl
2008-02-08 20:25         ` Harvey Harrison
2008-02-08 20:41           ` Jon Smirl
2008-02-09 15:27   ` Jan Holesovsky
2008-02-10  3:10     ` Nicolas Pitre
2008-02-10  4:59       ` Sean
2008-02-10  5:22         ` Nicolas Pitre
2008-02-10  5:35           ` Sean
2008-02-11  1:42             ` Jakub Narebski
2008-02-11  2:04               ` Nicolas Pitre
2008-02-11 10:11                 ` Jakub Narebski
2008-02-10  9:34         ` Joachim B Haga
2008-02-10 16:43       ` Johannes Schindelin
2008-02-10 17:01         ` Jon Smirl
2008-02-10 17:36           ` Johannes Schindelin
2008-02-10 18:47         ` Johannes Schindelin
2008-02-10 19:42           ` Nicolas Pitre
2008-02-10 20:11             ` Jon Smirl
2008-02-12 20:37           ` Johannes Schindelin
2008-02-12 21:05             ` Nicolas Pitre
2008-02-12 21:08             ` Linus Torvalds
2008-02-12 21:36               ` Jon Smirl
2008-02-12 21:59                 ` Linus Torvalds
2008-02-12 22:25                   ` Linus Torvalds
2008-02-12 22:43                     ` Jon Smirl
2008-02-12 23:39                       ` Linus Torvalds
2008-02-12 21:25             ` Jon Smirl
2008-02-14 19:20             ` Johannes Schindelin
2008-02-14 20:05               ` Jakub Narebski
2008-02-14 20:16                 ` Nicolas Pitre
2008-02-14 21:04                 ` Johannes Schindelin
2008-02-14 21:59                   ` Jakub Narebski
2008-02-14 23:38                     ` Johannes Schindelin
2008-02-14 23:51                       ` Brian Downing
2008-02-14 23:57                         ` Brian Downing
2008-02-15  0:08                         ` Johannes Schindelin
2008-02-15  1:41                           ` Nicolas Pitre
2008-02-17  8:18                             ` Shawn O. Pearce
2008-02-17  9:05                               ` Junio C Hamano
2008-02-17 18:44                               ` Nicolas Pitre
2008-02-15  1:07                       ` Jakub Narebski
2008-02-15  9:43                     ` Jan Holesovsky
2008-02-14 21:08                 ` Brandon Casey
2008-02-15  9:34               ` Jan Holesovsky
2008-02-10 19:50         ` Nicolas Pitre
2008-02-14 19:41           ` Brandon Casey [this message]
2008-02-14 19:58             ` Johannes Schindelin
2008-02-14 20:11             ` Nicolas Pitre
2008-02-11  1:20     ` Jakub Narebski
2008-02-08 20:16 ` Johannes Schindelin
2008-02-08 21:35   ` Jakub Narebski
2008-02-08 21:52     ` Johannes Schindelin
2008-02-08 22:03       ` Mike Hommey
2008-02-08 22:34         ` Johannes Schindelin
2008-02-08 22:50           ` Mike Hommey
2008-02-08 23:14             ` Johannes Schindelin
2008-02-08 23:38               ` Mike Hommey
2008-02-09 21:20                 ` Jan Hudec
2008-02-09 15:54       ` Jan Holesovsky

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

  List information: http://vger.kernel.org/majordomo-info.html

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=47B4996B.4000900@nrlssc.navy.mil \
    --to=casey@nrlssc.navy.mil \
    --cc=Johannes.Schindelin@gmx.de \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=jnareb@gmail.com \
    --cc=kendy@suse.cz \
    --cc=nico@cam.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.
Code repositories for project(s) associated with this public inbox

	https://80x24.org/mirrors/git.git

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 read-only IMAP folder(s) and NNTP newsgroup(s).