notes/plan for hpss performance work

[Martin Pool]

On  1 May 2008, Aaron Bentley <aaron at aaronbentley.com> wrote:
> Martin Pool wrote:
> >  * Perhaps surprisingly, graph operations are not showing up as
> > dominant, at least in the cases we did here: pushing just one
> > revision, and pushing all of history.
> 
> In my experience, graph operations are a significant portion of creating
> a branch of a new project.  For example, when creating a branch of
> Launchpad from scratch, Bazaar takes 20 minutes to determine that it
> should fetch everything.

Are you talking about the push or pull case?  Coming into an existing
repository or not?  Thanks for pointing it out anyhow.

(Maybe to avoid roundtrips trying to reproduce your exact case you could
send to Andrew and I a -Dhpss trace of this operation on your real
situation?)

We were measuring on bzr.dev; it's possible this is small enough that
it's in a different performance domain to larger ones.  For example,
sometimes it can read the whole index in one go.  Also, I think we had
just a single pack in the destination repositories and that too may not
be representative.

I feel a bit torn between doing more testing and fixing things that we
already know are problems and fairly easy to fix.  I guess they are
parallelizable because of the time it takes to review and for large
benchmarks to run.

-- 
Martin