Re: [PATCH] parse_object: try internal cache before reading object db

On Thu, Jan 05, 2012 at 01:35:50PM -0800, Junio C Hamano wrote:

quoted

For example, GitHub's alternates repository for git.git has
~120,000 refs, of which only ~3200 are unique. The time for
upload-pack to print its list of advertised refs dropped
from 3.4s to 0.76s.

Nice. I am more impressed by 120k/3.4 than 3.2k/0.76, though ;-)

Actually, we can do much better than that. Here are a few patches that
avoid parsing objects when possible. They drop the 3.4s to 2.0s. If you
combine them with the parse_object optimization, my 120K case drops to
around 0.68s.

I don't know if it is really that worth it on top of the parse_object
optimization. It's almost negligible for the normal case (though I get a
tiny speedup on my ~900-ref git.git repo), and a minor speedup on the
crazy alternates case. OTOH, if you had some totally insane ref
structure, like 120K _unique_ refs (which would probably imply that
you're making one ref per commit or something silly like that. But hey,
people have suggested it in the past), then it could be a big
improvement.

  [1/2]: upload-pack: avoid parsing objects during ref advertisement
  [2/2]: upload-pack: avoid parsing tag destinations

-Peff