aurora-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Maxim Khutornenko" <>
Subject Re: Review Request 31525: Improving NearestFit reporting accuracy.
Date Fri, 27 Feb 2015 21:05:54 GMT

This is an automatically generated e-mail. To reply, visit:

(Updated Feb. 27, 2015, 9:05 p.m.)

Review request for Aurora, Kevin Sweeney and Bill Farner.


Kevin's comments.

Bugs: AURORA-1148

Repository: aurora


Instead of relying on a fixed MAX_SCORE, vetoes are now graded by their relative weight (severity)
with a dedicated constraint mismatch ranked higest and an insufficient resource mismatch -
lowest. Together with "break early" rule in SchedulingFilter, this ensures "heavier" vetoes
are given proper attention in the NearestFit. This simplifies `NearestFit` logic while also
improving pending reason reporting accuracy and scheduling performance.

Diffs (updated)

  src/main/java/org/apache/aurora/scheduler/filter/ 3313bd2f9ed5a88375d88715dea5da1e9ad8b963

  src/main/java/org/apache/aurora/scheduler/filter/ a020ce50d578fba7f32b370f448a49eb1c284147

  src/main/java/org/apache/aurora/scheduler/metadata/ c3097e49c0f6588ea765aa4fab69dd35e3d90e8b

  src/test/java/org/apache/aurora/scheduler/filter/ b00668423a53a8cf6f4da3456bce3354f1fd2424

  src/test/java/org/apache/aurora/scheduler/metadata/ 78a236c0f9074692b67ce18e6e03f18fe4529e02



./gradlew -Pq build


Maxim Khutornenko

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message