On Tue, Feb 18, 2020 at 02:36:55PM -0500, Derrick Stolee wrote: > > I wondered why you chose to extend these tests instead of using > > p5600-partial-clone.sh, but I guess this script definitely creates > > the bitmap for the test. When I tested p5600-partial-clone.sh below, > > I manually repacked the Linux repo to have a bitmap: > > > > Test v2.25.0 HEAD > > ---------------------------------------------------------------------------- > > 5600.2: clone without blobs 79.81(111.34+11.35) 36.00(69.37+7.30) -54.9% > > 5600.3: checkout of result 45.56(114.59+4.81) 46.43(80.50+5.41) +1.9% > > > > Perhaps results for these tests would also be appropriate for your > > commit messages? > > And speaking of valuable performance tests to update: should we take > the loop of fetch tests in p5311-pack-bitmaps-fetch.sh and make > equivalent versions in p5600-partial-clone.sh? It would be good to > make sure that the incremental fetches also improve in this scenario. I don't think it would make sense to add them permanently, since p5600 doesn't necessarily have bitmaps in effect. But in any case, those tests are mostly about pack-objects realizing when we can use on-disk deltas due to the presence of bitmaps. If we're avoiding sending blobs, then that cuts out a huge chunk of opportunity for it to make any improvement (it might still have some impact because of trees, though). So I'd expect the effect to be muted. I dunno. It's true that I just hypothesized a result which we could confirm. But the perf tests are quite expensive to run (and p5311 is one of the more expensive ones). I'm not sure that repeating those tests combined with partial clones carries a lot of regression-testing value to merit adding that expense to all future runs. -Peff
