From: Luke Kenneth Casson Leighton <luke.leighton@gmail.com>
To: Jeff King <peff@peff.net>
Cc: gitzilla@gmail.com, Nguyen Thai Ngoc Duy <pclouds@gmail.com>,
git <git@vger.kernel.org>
Subject: Re: git pack/unpack over bittorrent - works!
Date: Thu, 2 Sep 2010 19:07:07 +0100 [thread overview]
Message-ID: <AANLkTi=kO9USQYoTLQZyCRrjCHWRtPtd4S5EuFk4-gPv@mail.gmail.com> (raw)
In-Reply-To: <20100902155810.GB14508@sigill.intra.peff.net>
On Thu, Sep 2, 2010 at 4:58 PM, Jeff King <peff@peff.net> wrote:
> pack-objects will reuse previously found deltas. So the deltas you have
> in your existing packs matter. The deltas you have in your existing
> packs depend on many things. At least:
>
> 1. Options you used when packing (e.g., --depth and --window).
>
> 2. Probably exactly _when_ you packed. You could find a good delta
> from A to B. Later, object C comes into existence, and would
> provide a better delta base for B. I don't think we will ever try A
> against C, unless --no-reuse-delta is set.
>
> You have a different pack than somebody who packed after A, B, and
> C all existed.
>
> In practice, this tends not to happen much because the best deltas
> are usually going backwards in time to a previous version. But it
> can happen.
jeff, thanks for explaining (and to nicolas, i see, since beginning this)
mrhmfffh. just been reading Documentation/technical/pack-heuristics.txt.
so... options include:
* writing an alternative "canonical" pack-object algorithm. i'm
inclined to select "git format-patch"! :) but that would be the lazy
way....
* taking the seeder's pack-objects as the "canonical" ones,
regardless. cacheing of the results would, sadly, be virtually
unavoidable, given the situation (multi-threading etc.)
* throw away bittorrent entirely as a transport mechanism.
* force-feed one peer to be "the" provider of a particular given
pack. doesn't matter whom you contact to _obtain_ a pack from, as
long as you solely and exclusively get the pack from that particular
peer.
* slight improvement / variation on the above: if two peers just
coincidentally happen to create or have the same pack (as can be shown
by having the same SHA-1, and/or by having the same data in their
cache) then ta-daaa, you have a file-sharing network for that
particular pack.
i think.... i think i miiight be able, hmmm... i believe it would be
possible to implement this last option by creating separate .torrents
for packs (one each!). by splitting things down, so that pack objects
are named as {ref}-{objref}-{SHA-1}.torrent and by providing a "top
level" torrent which contains the refs/heads/* and the associate
rev-list(s)... each set of rev-lists would have the SHA-1 of the
pack-object that happened to be created (and shared) at that
particular time, from that particular client: you then genuinely don't
give a stuff about who has what, it's all the same, and...
hmmm, i feel a modification to / deviation from the bittorrent
protocol coming on :)
l.
next prev parent reply other threads:[~2010-09-02 18:07 UTC|newest]
Thread overview: 88+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-09-01 14:36 git pack/unpack over bittorrent - works! Luke Kenneth Casson Leighton
2010-09-01 22:04 ` Nguyen Thai Ngoc Duy
2010-09-02 13:37 ` Luke Kenneth Casson Leighton
2010-09-02 13:53 ` Luke Kenneth Casson Leighton
2010-09-02 14:08 ` Ævar Arnfjörð Bjarmason
2010-09-02 15:33 ` A Large Angry SCM
2010-09-02 15:42 ` Luke Kenneth Casson Leighton
2010-09-02 15:51 ` Luke Kenneth Casson Leighton
2010-09-02 17:06 ` A Large Angry SCM
2010-09-02 15:58 ` Jeff King
2010-09-02 16:41 ` Nicolas Pitre
2010-09-02 17:09 ` A Large Angry SCM
2010-09-02 17:31 ` Nicolas Pitre
2010-09-02 19:17 ` Luke Kenneth Casson Leighton
2010-09-02 19:29 ` Shawn O. Pearce
2010-09-02 19:51 ` Luke Kenneth Casson Leighton
2010-09-02 20:06 ` Luke Kenneth Casson Leighton
2010-09-03 0:36 ` Nicolas Pitre
2010-09-03 10:34 ` Luke Kenneth Casson Leighton
2010-09-03 17:03 ` Junio C Hamano
2010-09-02 20:28 ` Brandon Casey
2010-09-02 20:48 ` Luke Kenneth Casson Leighton
2010-09-02 20:45 ` Jakub Narebski
2010-09-02 21:10 ` Luke Kenneth Casson Leighton
2010-09-02 21:19 ` Luke Kenneth Casson Leighton
2010-09-03 0:29 ` Nicolas Pitre
2010-09-03 2:48 ` Nguyen Thai Ngoc Duy
2010-09-03 10:55 ` Luke Kenneth Casson Leighton
2010-09-03 10:23 ` Luke Kenneth Casson Leighton
2010-09-03 10:54 ` Luke Kenneth Casson Leighton
2010-09-02 18:07 ` Luke Kenneth Casson Leighton [this message]
2010-09-02 18:23 ` Casey Dahlin
2010-09-02 16:58 ` A Large Angry SCM
2010-09-02 17:21 ` Nicolas Pitre
2010-09-02 19:41 ` Luke Kenneth Casson Leighton
2010-09-02 19:52 ` A Large Angry SCM
2010-09-02 23:09 ` Nicolas Pitre
2010-09-03 10:37 ` Theodore Tso
2010-09-03 11:04 ` Luke Kenneth Casson Leighton
2010-09-03 17:12 ` Junio C Hamano
2010-09-03 18:31 ` Ted Ts'o
2010-09-03 19:41 ` Nicolas Pitre
2010-09-03 21:11 ` Luke Kenneth Casson Leighton
2010-09-04 0:24 ` Nguyen Thai Ngoc Duy
2010-09-04 0:57 ` Nguyen Thai Ngoc Duy
2010-09-04 1:52 ` Artur Skawina
2010-09-04 4:39 ` Nicolas Pitre
2010-09-04 5:42 ` Artur Skawina
2010-09-04 6:13 ` Nicolas Pitre
2010-09-04 11:58 ` Luke Kenneth Casson Leighton
2010-09-04 13:14 ` Luke Kenneth Casson Leighton
2010-09-05 2:16 ` Nicolas Pitre
2010-09-05 18:05 ` Luke Kenneth Casson Leighton
2010-09-05 23:52 ` Nicolas Pitre
2010-09-06 13:23 ` Luke Kenneth Casson Leighton
2010-09-06 16:51 ` Nicolas Pitre
2010-09-06 22:33 ` Luke Kenneth Casson Leighton
2010-09-06 23:34 ` Junio C Hamano
2010-09-06 23:57 ` Nicolas Pitre
2010-09-07 0:17 ` Luke Kenneth Casson Leighton
2010-09-07 0:29 ` Luke Kenneth Casson Leighton
2010-09-04 13:42 ` Artur Skawina
[not found] ` <20100904155638.GA17606@pcpool00.mathematik.uni-freiburg.de>
2010-09-04 17:23 ` Artur Skawina
2010-09-04 18:46 ` Artur Skawina
2010-09-04 1:57 ` Theodore Tso
2010-09-04 5:23 ` Kyle Moffett
2010-09-04 11:46 ` Theodore Tso
2010-09-04 14:06 ` Luke Kenneth Casson Leighton
2010-09-05 1:32 ` Nicolas Pitre
2010-09-05 17:16 ` Luke Kenneth Casson Leighton
2010-09-04 5:40 ` Nicolas Pitre
2010-09-04 12:00 ` Theodore Tso
2010-09-04 12:44 ` Luke Kenneth Casson Leighton
2010-09-04 14:50 ` Luke Kenneth Casson Leighton
2010-09-04 18:14 ` Ted Ts'o
2010-09-04 20:00 ` Luke Kenneth Casson Leighton
2010-09-04 22:41 ` Ted Ts'o
2010-09-05 17:22 ` Luke Kenneth Casson Leighton
2010-09-04 20:20 ` Jakub Narebski
2010-09-04 20:47 ` Luke Kenneth Casson Leighton
2010-09-04 21:16 ` Jakub Narebski
2010-09-04 21:24 ` Luke Kenneth Casson Leighton
2010-09-04 22:47 ` Ted Ts'o
2010-09-05 1:43 ` Tomas Carnecky
2010-09-05 1:18 ` Nicolas Pitre
2010-09-05 17:25 ` Luke Kenneth Casson Leighton
2010-09-06 0:05 ` Nicolas Pitre
2010-09-04 12:33 ` Luke Kenneth Casson Leighton
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='AANLkTi=kO9USQYoTLQZyCRrjCHWRtPtd4S5EuFk4-gPv@mail.gmail.com' \
--to=luke.leighton@gmail.com \
--cc=git@vger.kernel.org \
--cc=gitzilla@gmail.com \
--cc=pclouds@gmail.com \
--cc=peff@peff.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).