Mailing-List: contact commits-help@subversion.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@subversion.apache.org
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
From: Apache subversion Wiki <commits@subversion.apache.org>
To: Apache subversion Wiki <commits@subversion.apache.org>
Date: Sun, 13 Jul 2014 18:24:56 -0000
Message-ID: <20140713182456.95243.3029@eos.apache.org>
Subject: 
 =?utf-8?q?=5BSubversion_Wiki=5D_Trivial_Update_of_=22MeasuringRepositoryP?=
 =?utf-8?q?erformance=22_by_StefanFuhrmann?=
Auto-Submitted: auto-generated

Dear Wiki user,

You have subscribed to a wiki page or wiki category on "Subversion Wiki" fo=
r change notification.

The "MeasuringRepositoryPerformance" page has been changed by StefanFuhrman=
n:
https://wiki.apache.org/subversion/MeasuringRepositoryPerformance?action=3D=
diff&rev1=3D1&rev2=3D2

Comment:
typo

  =

  With this [[https://svn.apache.org/repos/asf/subversion/trunk/tools/dev/b=
enchmarks/RepoPerf/copy_repo.py|python script]], you also get random data f=
iles of up to the specified size being written in between revisions.  Those=
 get deleted once the copy is complete. Depending on the OS / SAN heuristic=
s, this may mimic the presence of other repositories growing as the copied =
one does.  Moreover, multiple repositories get copied at once in a round-ro=
bin interleaved scheme, i.e. copy rev 0 of all repositories, then rev 1 of =
all repositories, etc.
  =

- The spacing parameter should be equal to the average revision data being =
written to other repos for each revision / packed shard copy round.  IOW so=
mething like (total size of "missing" repos) / (number of revs in copied re=
pos).  As that may result in too much transiently disk space being allocati=
on, you may use 128 here.  That plays well with typical RAID strip and OS p=
refetch sizes.
+ The spacing parameter should be equal to the average revision data being =
written to other repos for each revision / packed shard copy round.  IOW so=
mething like (total size of "missing" repos) / (number of revs in copied re=
pos).  As that may result in too much transiently disk space being allocate=
d, you may use 128 here.  That plays well with typical RAID strip and OS pr=
efetch sizes.
  =

  =3D=3D What to test =3D=3D
  Before you conduct performance tests, you need to decide upon the scenari=
os as well as what kind of conclusions you want to be able to draw from the=
 results.  For the latter, it is often useful to identify the components an=
d settings involved and to be able to isolate their contribution.