-
Notifications
You must be signed in to change notification settings - Fork 28.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[SPARK-1103] [WIP] Automatic garbage collection of RDD, shuffle and broadcast data #126
Commits on Feb 5, 2014
-
Added unpersist method to Broadcast.
Roman Pastukhov committedFeb 5, 2014 Configuration menu - View commit details
-
Copy full SHA for 1e752f1 - Browse repository at this point
Copy the full SHA 1e752f1View commit details
Commits on Feb 6, 2014
-
Fix for Broadcast unpersist patch.
Updated comment in MemoryStore.dropFromMemory Keep TorrentBroadcast piece blocks until unpersist is called
Roman Pastukhov committedFeb 6, 2014 Configuration menu - View commit details
-
Copy full SHA for 80dd977 - Browse repository at this point
Copy the full SHA 80dd977View commit details
Commits on Feb 14, 2014
-
Added ContextCleaner to automatically clean RDDs and shuffles when th…
…ey fall out of scope. Also replaced TimeStampedHashMap to BoundedHashMaps and TimeStampedWeakValueHashMap for the necessary hashmap behavior.
Configuration menu - View commit details
-
Copy full SHA for e427a9e - Browse repository at this point
Copy the full SHA e427a9eView commit details -
Configuration menu - View commit details
-
Copy full SHA for 8512612 - Browse repository at this point
Copy the full SHA 8512612View commit details
Commits on Mar 11, 2014
-
Merge remote-tracking branch 'apache/master' into state-cleanup
Conflicts: core/src/main/scala/org/apache/spark/MapOutputTracker.scala core/src/main/scala/org/apache/spark/SparkContext.scala core/src/main/scala/org/apache/spark/scheduler/ResultTask.scala core/src/main/scala/org/apache/spark/scheduler/ShuffleMapTask.scala core/src/main/scala/org/apache/spark/storage/BlockManager.scala core/src/main/scala/org/apache/spark/util/TimeStampedHashMap.scala core/src/test/scala/org/apache/spark/MapOutputTrackerSuite.scala
Configuration menu - View commit details
-
Copy full SHA for a24fefc - Browse repository at this point
Copy the full SHA a24fefcView commit details -
Configuration menu - View commit details
-
Copy full SHA for cb0a5a6 - Browse repository at this point
Copy the full SHA cb0a5a6View commit details
Commits on Mar 12, 2014
-
Removed unncessary TimeStampedHashMap from DAGScheduler, added try-ca…
…tches in finalize() methods, and replaced ArrayBlockingQueue to LinkedBlockingQueue to avoid blocking in Java's finalizing thread.
Configuration menu - View commit details
-
Copy full SHA for ae9da88 - Browse repository at this point
Copy the full SHA ae9da88View commit details
Commits on Mar 13, 2014
-
Configuration menu - View commit details
-
Copy full SHA for e61daa0 - Browse repository at this point
Copy the full SHA e61daa0View commit details
Commits on Mar 17, 2014
-
Added try-catch in context cleaner and null value cleaning in TimeSta…
…mpedWeakValueHashMap.
Configuration menu - View commit details
-
Copy full SHA for a7260d3 - Browse repository at this point
Copy the full SHA a7260d3View commit details
Commits on Mar 18, 2014
-
Removed use of BoundedHashMap, and made BlockManagerSlaveActor cleanu…
…p shuffle metadata in MapOutputTrackerWorker.
Configuration menu - View commit details
-
Copy full SHA for 892b952 - Browse repository at this point
Copy the full SHA 892b952View commit details
Commits on Mar 19, 2014
-
Configuration menu - View commit details
-
Copy full SHA for e1fba5f - Browse repository at this point
Copy the full SHA e1fba5fView commit details
Commits on Mar 25, 2014
-
Configuration menu - View commit details
-
Copy full SHA for f2881fd - Browse repository at this point
Copy the full SHA f2881fdView commit details -
Configuration menu - View commit details
-
Copy full SHA for 620eca3 - Browse repository at this point
Copy the full SHA 620eca3View commit details -
Merge remote-tracking branch 'apache/master' into state-cleanup
Conflicts: core/src/main/scala/org/apache/spark/Dependency.scala core/src/main/scala/org/apache/spark/MapOutputTracker.scala core/src/main/scala/org/apache/spark/SparkContext.scala core/src/main/scala/org/apache/spark/SparkEnv.scala core/src/main/scala/org/apache/spark/rdd/RDD.scala core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala core/src/main/scala/org/apache/spark/scheduler/ShuffleMapTask.scala core/src/main/scala/org/apache/spark/storage/BlockManager.scala core/src/main/scala/org/apache/spark/storage/ThreadingTest.scala core/src/test/scala/org/apache/spark/storage/BlockManagerSuite.scala
Configuration menu - View commit details
-
Copy full SHA for a007307 - Browse repository at this point
Copy the full SHA a007307View commit details -
Configuration menu - View commit details
-
Copy full SHA for d2f8b97 - Browse repository at this point
Copy the full SHA d2f8b97View commit details -
Configuration menu - View commit details
-
Copy full SHA for 6c9dcf6 - Browse repository at this point
Copy the full SHA 6c9dcf6View commit details
Commits on Mar 26, 2014
-
Merge branch 'bc-unpersist-merge' of github.com:ignatich/incubator-sp…
…ark into cleanup Conflicts: core/src/main/scala/org/apache/spark/broadcast/BroadcastFactory.scala core/src/main/scala/org/apache/spark/broadcast/TorrentBroadcast.scala core/src/main/scala/org/apache/spark/storage/MemoryStore.scala
Configuration menu - View commit details
-
Copy full SHA for c7ccef1 - Browse repository at this point
Copy the full SHA c7ccef1View commit details -
Configuration menu - View commit details
-
Copy full SHA for ba52e00 - Browse repository at this point
Copy the full SHA ba52e00View commit details -
Add framework for broadcast cleanup
As of this commit, Spark does not clean up broadcast blocks. This will be done in the next commit.
Configuration menu - View commit details
-
Copy full SHA for d0edef3 - Browse repository at this point
Copy the full SHA d0edef3View commit details -
Configuration menu - View commit details
-
Copy full SHA for 544ac86 - Browse repository at this point
Copy the full SHA 544ac86View commit details
Commits on Mar 27, 2014
-
Add tests for unpersisting broadcast
There is not currently a way to query the blocks on the executors, an operation that is deceptively simple to accomplish. This commit adds this mechanism in order to verify that blocks are in fact persisted/unpersisted on the executors in the tests.
Configuration menu - View commit details
-
Copy full SHA for e95479c - Browse repository at this point
Copy the full SHA e95479cView commit details -
Configuration menu - View commit details
-
Copy full SHA for f201a8d - Browse repository at this point
Copy the full SHA f201a8dView commit details -
Merge github.com:apache/spark into cleanup
Conflicts: core/src/main/scala/org/apache/spark/SparkContext.scala
Configuration menu - View commit details
-
Copy full SHA for c92e4d9 - Browse repository at this point
Copy the full SHA c92e4d9View commit details
Commits on Mar 28, 2014
-
Configuration menu - View commit details
-
Copy full SHA for 0d17060 - Browse repository at this point
Copy the full SHA 0d17060View commit details -
Generalize BroadcastBlockId to remove BroadcastHelperBlockId
Rather than having a special purpose BroadcastHelperBlockId just for TorrentBroadcast, we now have a single BroadcastBlockId that has a possibly empty field. This simplifies broadcast clean-up because now we only have to look for one type of block. This commit also simplifies BlockId JSON de/serialization in general by parsing the name through regex with apply.
Configuration menu - View commit details
-
Copy full SHA for 34f436f - Browse repository at this point
Copy the full SHA 34f436fView commit details
Commits on Mar 29, 2014
-
Add functionality to query executors for their local BlockStatuses
Not all blocks are reported to the master. In HttpBroadcast and TorrentBroadcast, for instance, most blocks are not reported to master. The lack of a mechanism to get local block statuses on each executor makes it difficult to test the correctness of un/persisting a broadcast. This new functionality, though only used for testing at the moment, is general enough to be used for other things in the future.
Configuration menu - View commit details
-
Copy full SHA for fbfeec8 - Browse repository at this point
Copy the full SHA fbfeec8View commit details -
Make TimeStampedWeakValueHashMap a wrapper of TimeStampedHashMap
This allows us to get rid of WrappedJavaHashMap without much duplicate code.
Configuration menu - View commit details
-
Copy full SHA for 88904a3 - Browse repository at this point
Copy the full SHA 88904a3View commit details -
Configuration menu - View commit details
-
Copy full SHA for e442246 - Browse repository at this point
Copy the full SHA e442246View commit details
Commits on Mar 30, 2014
-
Configuration menu - View commit details
-
Copy full SHA for 8557c12 - Browse repository at this point
Copy the full SHA 8557c12View commit details
Commits on Mar 31, 2014
-
Configuration menu - View commit details
-
Copy full SHA for 7edbc98 - Browse repository at this point
Copy the full SHA 7edbc98View commit details -
Configuration menu - View commit details
-
Copy full SHA for 634a097 - Browse repository at this point
Copy the full SHA 634a097View commit details -
Configuration menu - View commit details
-
Copy full SHA for 7ed72fb - Browse repository at this point
Copy the full SHA 7ed72fbView commit details
Commits on Apr 1, 2014
-
Configuration menu - View commit details
-
Copy full SHA for 5016375 - Browse repository at this point
Copy the full SHA 5016375View commit details
Commits on Apr 2, 2014
-
Correct semantics for TimeStampedWeakValueHashMap + add tests
This largely accounts for the cases when WeakReference becomes no longer strongly reachable, in which case the map should return None for all get() operations, and should skip the entry for all listing operations.
Configuration menu - View commit details
-
Copy full SHA for f0aabb1 - Browse repository at this point
Copy the full SHA f0aabb1View commit details -
Merge pull request #1 from andrewor14/cleanup
I am merging this. I will take one more detailed look in the context of my original changes in the main PR.
Configuration menu - View commit details
-
Copy full SHA for 762a4d8 - Browse repository at this point
Copy the full SHA 762a4d8View commit details
Commits on Apr 4, 2014
-
Merge github.com:apache/spark into cleanup
Conflicts: core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala core/src/main/scala/org/apache/spark/util/JsonProtocol.scala
Configuration menu - View commit details
-
Copy full SHA for a6460d4 - Browse repository at this point
Copy the full SHA a6460d4View commit details -
Configuration menu - View commit details
-
Copy full SHA for c5b1d98 - Browse repository at this point
Copy the full SHA c5b1d98View commit details -
Merge remote-tracking branch 'apache/master' into state-cleanup
Conflicts: core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala core/src/main/scala/org/apache/spark/util/JsonProtocol.scala
Configuration menu - View commit details
-
Copy full SHA for a2cc8bc - Browse repository at this point
Copy the full SHA a2cc8bcView commit details -
Configuration menu - View commit details
-
Copy full SHA for ada45f0 - Browse repository at this point
Copy the full SHA ada45f0View commit details -
Configuration menu - View commit details
-
Copy full SHA for cd72d19 - Browse repository at this point
Copy the full SHA cd72d19View commit details -
Configuration menu - View commit details
-
Copy full SHA for b27f8e8 - Browse repository at this point
Copy the full SHA b27f8e8View commit details -
Configuration menu - View commit details
-
Copy full SHA for a430f06 - Browse repository at this point
Copy the full SHA a430f06View commit details -
Fixed failing BroadcastSuite unit tests by introducing blocking for r…
…emoveShuffle and removeBroadcast in BlockManager*
Configuration menu - View commit details
-
Copy full SHA for 104a89a - Browse repository at this point
Copy the full SHA 104a89aView commit details -
Configuration menu - View commit details
-
Copy full SHA for 6222697 - Browse repository at this point
Copy the full SHA 6222697View commit details
Commits on Apr 7, 2014
-
Configuration menu - View commit details
-
Copy full SHA for 41c9ece - Browse repository at this point
Copy the full SHA 41c9eceView commit details -
Added more documentation on Broadcast implementations, specially whic…
…h blocks are told about to the driver. Also, fixed Broadcast API to hide destroy functionality.
Configuration menu - View commit details
-
Copy full SHA for 2b95b5e - Browse repository at this point
Copy the full SHA 2b95b5eView commit details -
Configuration menu - View commit details
-
Copy full SHA for 4d05314 - Browse repository at this point
Copy the full SHA 4d05314View commit details -
Configuration menu - View commit details
-
Copy full SHA for cff023c - Browse repository at this point
Copy the full SHA cff023cView commit details -
Configuration menu - View commit details
-
Copy full SHA for d25a86e - Browse repository at this point
Copy the full SHA d25a86eView commit details
Commits on Apr 8, 2014
-
Merge remote-tracking branch 'apache/master' into state-cleanup
Conflicts: core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala core/src/main/scala/org/apache/spark/storage/BlockManager.scala core/src/main/scala/org/apache/spark/storage/BlockManagerMasterActor.scala core/src/main/scala/org/apache/spark/storage/BlockManagerMessages.scala
Configuration menu - View commit details
-
Copy full SHA for f489fdc - Browse repository at this point
Copy the full SHA f489fdcView commit details -
Configuration menu - View commit details
-
Copy full SHA for 61b8d6e - Browse repository at this point
Copy the full SHA 61b8d6eView commit details