Skip to content

tryout fsst decompress speedup #6998

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
joseph-isaacs wants to merge 4 commits into
base: develop
Choose a base branch
from
+96 −108

fix

a05fcab
Select commit
Loading
Failed to load commit list.
Draft

tryout fsst decompress speedup #6998

fix
a05fcab
Select commit
Loading
Failed to load commit list.
CodSpeed HQ / CodSpeed Performance Analysis succeeded Mar 17, 2026 in 0s

Performance Gate Passed

⚡ 21 improved benchmarks
✅ 988 untouched benchmarks
🆕 20 new benchmarks
⏩ 1515 skipped benchmarks1

Performance Changes

Mode Benchmark BASE HEAD Efficiency
🆕 Simulation canonicalize_compare[(10000, 256, 8)] N/A 10.4 ms N/A
🆕 Simulation canonicalize_compare[(1000, 256, 4)] N/A 944.7 µs N/A
🆕 Simulation canonicalize_compare[(1000, 256, 8)] N/A 1.1 ms N/A
⚡ Simulation canonicalize_compare[(10000, 16, 8)] 1.5 ms 1.4 ms +11.08%
⚡ Simulation canonicalize_compare[(10000, 16, 4)] 1.4 ms 1.3 ms +12.02%
⚡ Simulation canonicalize_compare[(10000, 4, 4)] 1.2 ms 1 ms +13.93%
🆕 Simulation canonicalize_compare[(10000, 256, 4)] N/A 8.6 ms N/A
⚡ Simulation chunked_into_canonical[(1000, 100, 16, 4)] 15.2 ms 13.6 ms +11.94%
⚡ Simulation chunked_into_canonical[(10, 10000, 16, 4)] 11.6 ms 10.1 ms +15.13%
⚡ Simulation chunked_canonicalize_into[(10, 10000, 4, 4)] 7.8 ms 6.4 ms +21.79%
🆕 Simulation compress_fsst[(1000, 256, 8)] N/A 3.5 ms N/A
⚡ Simulation chunked_into_canonical[(10, 10000, 4, 4)] 7.8 ms 6.4 ms +21.75%
⚡ Simulation chunked_canonicalize_into[(1000, 100, 16, 4)] 15 ms 13.4 ms +12.12%
⚡ Simulation chunked_into_canonical[(1000, 50, 8, 4)] 8 ms 7.3 ms +10.81%
⚡ Simulation chunked_canonicalize_into[(1000, 50, 8, 4)] 7.8 ms 7 ms +11.2%
⚡ Simulation canonicalize_compare[(10000, 4, 8)] 1.2 ms 1 ms +13.57%
🆕 Simulation compress_fsst[(1000, 256, 4)] N/A 2.9 ms N/A
⚡ Simulation chunked_canonicalize_into[(10, 10000, 16, 4)] 11.6 ms 10 ms +15.11%
🆕 Simulation compress_fsst[(10000, 256, 4)] N/A 28.7 ms N/A
🆕 Simulation compress_fsst[(10000, 256, 8)] N/A 34.1 ms N/A
... ... ... ... ... ...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

Comparing ji/make-view (a05fcab) with develop (50868db)

Open in CodSpeed

Footnotes

  1. 1515 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

View more details on CodSpeed HQ