itlxrd/psysonic
1
0
Fork 0
mirror of https://github.com/kilyabin/psysonic.git synced 2026-07-21 22:15:40 +00:00
Code Issues Packages Projects Releases Wiki Activity

perf(analysis): measure tpm over trailing 5s window (#948)

Browse Source 
* perf(analysis): measure tpm over trailing 5s instead of full minute

Mirror the cover cpm change: a 60s rolling average added too much inertia and
flattened real bursts/stalls. Count completions in the trailing 5s window and
extrapolate to per-minute, so analysis tpm reacts promptly and decays to 0
within the window when idle. Retention stays at 60s.

* docs(changelog): note trailing-window throughput rate (PR #948)
This commit is contained in:
cucadmuh
2026-06-02 12:17:10 +03:00
committed by GitHub
parent 2224ddbe78
commit 81f900c7a6
4 changed files with 43 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files
CHANGELOG.md
+7
View File
@@ -388,6 +388,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
### Performance Probe — responsive throughput windows (tpm / cpm)
**By [@cucadmuh](https://github.com/cucadmuh), PR [#948](https://github.com/Psychotoxical/psysonic/pull/948)**
* Analysis **tpm** and cover **cpm** (lib + ui) now measure throughput over the trailing **5 seconds** instead of a full-minute rolling average. The figure is still extrapolated to per-minute, but reacts promptly to bursts/stalls and decays to 0 within the window when idle, instead of coasting on minute-long inertia.
## Fixed