aboutsummaryrefslogtreecommitdiff
path: root/docs
diff options
context:
space:
mode:
authorAndrew Or <andrew@databricks.com>2015-12-01 19:36:34 -0800
committerJosh Rosen <joshrosen@databricks.com>2015-12-01 19:36:34 -0800
commit1ce4adf55b535518c2e63917a827fac1f2df4e8e (patch)
treee4ca6006d6a4f2e30f6d320e03bd3c0f7f7688ce /docs
parente96a70d5ab2e2b43a2df17a550fa9ed2ee0001c4 (diff)
downloadspark-1ce4adf55b535518c2e63917a827fac1f2df4e8e.tar.gz
spark-1ce4adf55b535518c2e63917a827fac1f2df4e8e.tar.bz2
spark-1ce4adf55b535518c2e63917a827fac1f2df4e8e.zip
[SPARK-8414] Ensure context cleaner periodic cleanups
Garbage collection triggers cleanups. If the driver JVM is huge and there is little memory pressure, we may never clean up shuffle files on executors. This is a problem for long-running applications (e.g. streaming). Author: Andrew Or <andrew@databricks.com> Closes #10070 from andrewor14/periodic-gc.
Diffstat (limited to 'docs')
0 files changed, 0 insertions, 0 deletions