grit://cooking/bench/remote/remove@many-refs
← cookingall benchmarksraw data

grit remote · remove @many-refs

Remove origin with 10k packed tracking refs vs git remote remove

repo: 10,000 packed refs · 200 files · 1,000 commits · 10,000 refs

$ grit-bench daily --only remote/remove@many-refs2026-10-27 · grit 8dbc58e77 · git version 2.43.0
toolmedianminmaxrunspeak memory
git745 ms735 ms756 ms209.6 MB
grit338 ms332 ms357 ms2025 MB
grit is 2.2× fastermemory: 2.6× more
every timed run (log scale)
git
grit
264 ms952 ms
How it's measured
setup
remote http://127.0.0.1:43971/remote.git served by git http-backend (git's own smart HTTP server)
before each run
set -eu 'grit-bench' prepare-daily-reset --git git cd /tmp/grit-bench-scratch/daily-remote-many-refs-ls/remote.git-campaign-push-client git remote remove origin 2>/dev/null || true git remote add origin 'http://127.0.0.1:43971/remote.git' git fetch -q origin '+refs/heads/*:refs/remotes/origin/*' git pack-refs '--all'
git runs
$ git remote remove origin
grit runs
$ grit remote remove origin
Where the time goesone extra traced run per tool, outside the timed runs
git
— not traced
grit · median run 338 ms
grit.repository.open
0.3 ms
grit.config.load ×3
0.2 ms
Every day8 days
dategrit commitgitgritgrit vs gitgrit peak
2026-10-278dbc58e77745 ms338 ms2.2× faster25 MB
2026-10-26c7cf2c443621 ms764 ms1.2× slower25 MB
2026-10-252824ad391866 ms396 ms2.2× faster25 MB
2026-10-24157a05dde835 ms405 ms2.1× faster24 MB
2026-10-23bddf92f86760 ms29.60 s39× slower29 MB
2026-10-224c131b942—not run——
2026-10-21d4844e92a—not run——
2026-10-204754ad0b9—not run——