4paradigm / 4paradigm/OpenMLDB
online performance with long window optimization degrades after upgrade
Offen
bug
- Vorherrschende Sprache
- C++
- Sterne
- 1.7k
- Forks
- 331
- Ø Merge
- 12 T. 12 Std.
- Gemergte PRs (30 T.)
- 1
Beschreibung
**Bug Description**
- online performance with long window optimization degrades after upgrade
- after re-deployment, the performance looks normal
**Expected Behavior**
**Relation Case**
**Steps to Reproduce**
1.
2.
3.
4.
Beitragsleitfaden
Rechercherichtung
The issue mentions 'long window optimization' and performance degradation after an upgrade. Start by examining recent changes to the windowing logic in the codebase, particularly around performance-critical paths. Look for any related tests or benchmarks that measure online performance. The goal is to identify what changed in the upgrade that caused the regression and verify that a re-deployment temporarily fixes it.
Vom Indexierungsmodell aus dem Issue-Text verfasst.
Bewertung
- Tech-Stack
- databases, machine-learning
- Bereich
- databases, machine-learning, performance
- Issue-Typ
- Bug
- Schwierigkeit
- 4/5
- Geschätzter Aufwand
- 3-5 Tage
- Aktivitätsstatus
- Veraltet
- Klarheit
- Muss geklärt werden
- Anfängerfreundlichkeit
- 25/100