Gentoo Archives: gentoo-scm

From: Donnie Berkholz <dberkholz@g.o>
To: "Robin H. Johnson" <robbat2@g.o>
Cc: gentoo-scm@l.g.o
Subject: Re: [gentoo-scm] Testing git conversion
Date: Tue, 28 Oct 2008 22:36:35
Message-Id: 20081028223630.GA19467@comet
In Reply to: Re: [gentoo-scm] Testing git conversion by "Robin H. Johnson"
On 01:11 Tue 28 Oct     , Robin H. Johnson wrote:
> On Tue, Oct 28, 2008 at 07:21:10AM +0000, Ciaran McCreesh wrote: > > More likely you're just having problems with the connection lasting for > > too long. 'git fetch' will resume (on a per-object basis, which is > > fine so long as there aren't too many huge packs). > > It's a single ~910MiB pack. I'll play with it and see how we can make it > better, maybe limiting the pack sizes to 50Mb/ea.
I did a few repacks and found that there's a linear relationship between the log of the pack size and the repo size. If repo size is in GB and pack size is in MB: repo_size=m*(log(pack_size))+c, where m=-0.35, c=3.3 Using that, it's straightforward to optimize on whichever parameter you care about while still considering the other one. I also attached two graphs showing this more visually, one in log form and the other not. The clear preference is the largest pack size where we rarely encounter clone problems. Perhaps we should just cut the pack into fractions (900/2, 900/3, etc) until we no longer hear about problems cloning. One major issue is that apparently partial downloads of a single pack don't resume. Ideally, we should just get that fixed upstream and use a single huge pack. Thoughts? -- Thanks, Donnie Donnie Berkholz Developer, Gentoo Linux Blog:


File name MIME type
pack-size-log.png image/png
pack-size.png image/png


Subject Author
Re: [gentoo-scm] Testing git conversion Donnie Berkholz <dberkholz@g.o>
Re: [gentoo-scm] Testing git conversion Christian Faulhammer <opfer@g.o>