diff options
| author | Lukasz Kasprzak <lukas@labunix.xyz> | 2026-09-12 21:34:22 +0200 |
|---|---|---|
| committer | Lukasz Kasprzak <lukas@labunix.xyz> | 2026-09-12 21:34:22 +0200 |
| commit | a6e442a645902011b2081c216daaec052cdc6ce6 (patch) | |
| tree | 9af223981b78f3ebfcc498958c34b2cdb22c0fcf /docs | |
| parent | 3f8679be9373ee7508d512dfdfc1dda0839c7f90 (diff) | |
| download | krino-a6e442a645902011b2081c216daaec052cdc6ce6.tar.gz krino-a6e442a645902011b2081c216daaec052cdc6ce6.zip | |
krino: no performance target for 0.0.1
Lukasz: "no performance goal". Section 13s "under 2 s on a 4-core laptop" is
replaced with the decision and its evidence, so it is not reinstated later:
measured throughput is dominated by the external extractors, not by krino. A
real folder of 265 files with 164 needing pdftotext at roughly 80 ms each takes
12-17 s; a synthetic 4405-file tree needing no extraction takes 1.09 s. A single
threshold would describe popplers speed and the shape of one folder. The
measurement is still reported in the release notes; nothing is promised.
Diffstat (limited to 'docs')
| -rw-r--r-- | docs/design.md | 10 |
1 files changed, 8 insertions, 2 deletions
diff --git a/docs/design.md b/docs/design.md index 42c691c..8e4e914 100644 --- a/docs/design.md +++ b/docs/design.md @@ -524,8 +524,14 @@ Rules for the GUI to come: keyword matching matters next to extraction. - Duplicates: stat-only grouping, partial hash, then full hash. - Measured before release on a real downloads directory of a few hundred - files, about half of them PDFs, and reported in the release notes. Goal: a - full plan with content rules in under 2 s on a 4-core laptop. `make bench` + files, about half of them PDFs, and reported in the release notes. + **No performance target for 0.0.1.** Measured throughput is dominated by the + external extractors, not by krino: on a real folder of 265 files, 164 of them + needing `pdftotext` at roughly 80 ms each, a full plan takes 12-17 s, while a + synthetic tree of 4405 files needing no extraction takes 1.09 s. A single + threshold would therefore describe poppler's speed and the shape of one + folder rather than anything about krino, and would fail or pass for reasons + outside its control. Report the measurement; promise nothing. `make bench` runs the Go benchmarks on generated trees. ## 14. Build, dependencies, release |