David Glasser wrote:
> On Sat, Oct 25, 2008 at 11:45 PM, David Glasser
> <glasser_at_davidglasser.net> wrote:
>> On Sat, Oct 25, 2008 at 3:10 PM, C. Michael Pilato <cmpilato_at_collab.net> wrote:
>>> I did a little bit of simple comparison testing between FSFS and BDB in
>>> 1.5.4 and trunk. My testing involved loading a dumpfile of 5000 revisions
>>> (taken from our own repository), then doing some single-revision dumps at a
>>> few time-slices across the loaded repository.
>>>
>>> Here are the highlights (percentages are ballpark estimates).
>>>
>>> In trunk, FSFS:
>>>
>>> is significantly slower (30%) for writes operations. I have no idea why.
>> As discussed in the other thread, the obvious thing to assume would be
>> that all the reads and rights of the rep cache DB are slowing you
>> down... try disabling that (just take out all the calls into
>> rep-cache.h from fs-fs.c) and try again?
>
> Have you had a chance to try this? I can give you the appropriate
> patch if you'd like.
I had a chance to try this earlier today, but don't have the numbers handy right
now. I only did FSFS experiments, but the basic gist was that trunk without
rep-sharing was faster, but not quite as fast as 1.5.x.
I'm working on a patch which allows rep-sharing to be configurable, via
fsfs.conf. This should make playing with these results a bit easier.
-Hyrum
Received on 2008-10-29 23:42:31 CET