spark-gc-metrics-hack

public

0 stars

1 forks

1 issues

Commits

List of commits on branch master.

Unverified

dbd9f39a255907b921d5d777fe082493bd562027

Merge branch 'master' of github.com:woggle/spark-gc-metrics-hack

ccharlesreiss committed 7 years ago

Unverified

d47d1be55abb236488a4659500e6ba60d949b483

more GC sampling policy support

ccharlesreiss committed 7 years ago

Unverified

d97809a347ae793bdf2a790c9549d9ec2ec7fea8

Add README.

ccharlesreiss committed 9 years ago

Unverified

c220abb829d64ad8bed49425c40a8d1b1f5198ea

Instrument have-native-alloc and sample half and native alloc

ccharlesreiss committed 10 years ago

Unverified

dd18376427e2ebf7f0ed72d09d95c7e6a672adb3

gcVary.sampleHalf

ccharlesreiss committed 10 years ago

Unverified

50a5bed5083140b4b63ae30a54544b4cb9a5387d

JNITricks working; extra old-for-young fields

ccharlesreiss committed 11 years ago

README

The README file for this repository.

Hackish tools for logging a whole lot of GC statistics from Spark and logging them in a machine-readable JSON format after every GC.

Also, can tie in with a modified JVM that supports making "dummy" heap allocations in an efficient way. This allows us to vary the effective size of the eden space without "real" support for this.