[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Subversion 1.9.0-dev FSFS performance tests

From: Stefan Fuhrmann <stefan.fuhrmann_at_wandisco.com>
Date: Fri, 20 Jun 2014 01:55:20 +0200

On Thu, Jun 19, 2014 at 11:38 PM, Branko Čibej <brane_at_wandisco.com> wrote:

> On 19.06.2014 17:06, Stefan Fuhrmann wrote:
>
> Turn out that the ruby repo is something special
> in that it has very deep histories of relatively few,
> very small files combined with one huge changelog
> file (the latter taking up ~75% of the repo). See
> below for details.
>
> Also, please note that your exports contained
> >500000 files. Using 16MB of cache with that
> project size *may* not be an adequate setup.
> Upping that to insane 256MB (roughly what 1.6
> would use anyway), gives much better numbers.
> However, there is hardly a difference between
> f6 and f7 in these runs.
>
>
>
> Heh, this sound suspiciously like saying that one has to have the right
> test data to make v7 faster than v6. :)
>

Sure. As far as I can see, we always end up reading
most of the repository in this specific case as there
are virtually no entirely cool data blocks. It's only ~40
pack files and most files seem to be present since early
in the project -> most paths need to be read from almost
all pack files.

Since the whole repo is also quite small on disk (~400MB),
everything ends up in disk cache and the reading order
is of little importance. The BSD repo OTOH, has 20x
as much data in their "trunk" ("/head") and the history
is 5x deeper. Plus lots of branching and merging and
f7 reorg-on-pack helps quite a bit.

-- Stefan^2.
Received on 2014-06-20 01:55:48 CEST

This is an archived mail posted to the Subversion Dev mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.