grit://cooking/bench/switch/large-blob@big-asset
← cookingall benchmarksraw data

grit switch · large-blob @big-asset

Switch away from a 300 MB loose text blob and back, writing it to the work tree: grit switch vs git switch

repo: Large loose text asset, plus README; other branch deletes the asset · 2 files · 2 commits · 3 refs

$ grit-bench daily --only switch/large-blob@big-asset2026-10-27 · grit 8dbc58e77 · git version 2.43.0
toolmedianminmaxrunspeak memory
git828 ms814 ms1.43 s20307 MB
grit738 ms493 ms767 ms2019 MB
grit is 1.1× fastermemory: 16× less
every timed run (log scale)
git
grit
391 ms1.80 s
How it's measured
before each run
grit-bench prepare-switch --git git
git runs
$ git switch -q bench-b && git switch -q bench-a
grit runs
$ grit switch bench-b && grit switch bench-a
Where the time goesone extra traced run per tool, outside the timed runs
git · median run 828 ms
unpack_trees: unpack_trees ×2
821 ms
index: do_write_index ×2
0.1 ms
cache_tree: update ×2
0.1 ms
index: do_read_index ×2
0.1 ms
unpack_trees: traverse_trees ×2
0.0 ms
index: refresh ×2
0.0 ms
cache_tree: read ×2
0.0 ms
cache_tree: write ×2
0.0 ms
grit · median run 738 ms
grit.checkout.write
463 ms
grit.repository.open ×2
0.4 ms
grit.config.load ×6
0.2 ms
grit.index.write ×2
0.2 ms
grit.index.read ×2
0.1 ms
grit.checkout.plan ×2
0.1 ms
Every day8 days
dategrit commitgitgritgrit vs gitgrit peak
2026-10-278dbc58e77828 ms738 ms1.1× faster19 MB
2026-10-26c7cf2c443842 ms482 ms1.7× faster18 MB
2026-10-252824ad391873 ms809 ms1.1× faster17 MB
2026-10-24157a05dde—not run——
2026-10-23bddf92f86—not run——
2026-10-224c131b942—not run——
2026-10-21d4844e92a—not run——
2026-10-204754ad0b9—not run——