ODIn/spark-instrumented-optimizer
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity
3210 commits 6 branches 181 tags 397 MiB
Commit graph

3210 commits

This branch
This branch
All branches
Author SHA1 Message Date
Charles Reiss 88b9d240fd Remove dead code in test. 2013-01-23 12:40:38 -08:00
Matei Zaharia 1a3aeeca23 Merge pull request #407 from woggling/no-cache-tracker 
Eliminate CacheTracker
 2013-01-23 12:28:48 -08:00
Charles Reiss e1027ca639 Actually add CacheManager. 2013-01-23 12:22:11 -08:00
Matei Zaharia 4147e1d47b Merge pull request #406 from tdas/master 
Changed StorageLevel and BlockManagerId API to prevent duplication in memory
 2013-01-23 12:18:31 -08:00
Matei Zaharia 4d77d554e1 Merge pull request #394 from JoshRosen/add_file_fix 
Add SparkFiles.get() API to access files added through addFile().
 2013-01-23 12:16:30 -08:00
Josh Rosen b47d054cfc Remove use of abc.ABCMeta due to cloudpickle issue. 
cloudpickle runs into issues while pickling subclasses of AccumulatorParam,
which may be related to this Python issue:

    http://bugs.python.org/issue7689

This seems hard to fix and the ABCMeta wasn't necessary, so I removed it.
 2013-01-23 11:47:27 -08:00
Josh Rosen ae2ed2947d Allow PySpark's SparkFiles to be used from driver 
Fix minor documentation formatting issues.
 2013-01-23 10:58:50 -08:00
Tathagata Das 9c8ff1e55f Fixed checkpoint testcases 2013-01-23 07:31:49 -08:00
Tathagata Das 666ce431aa Added support for rescheduling unprocessed batches on master failure. 2013-01-23 03:15:36 -08:00
Tathagata Das 79d55700ce One more fix. Made even default constructor of BlockManagerId private to prevent such problems in the future. 2013-01-23 01:57:09 -08:00
Charles Reiss 0b506dd2ec Add tests of various node failure scenarios. 2013-01-23 01:38:15 -08:00
Charles Reiss d209b6b764 Extra debugging from hostLost() 2013-01-23 01:35:14 -08:00
Charles Reiss 9a27062260 Force generation increment after shuffle map stage 2013-01-23 01:34:44 -08:00
Tathagata Das 155f31398d Made StorageLevel constructor private, and added StorageLevels.create() to the Java API. Updates scala and java programming guides. 2013-01-23 01:10:26 -08:00
Tathagata Das 5e11f1e51f Modified StorageLevel API to ensure zero duplicate objects. 2013-01-22 23:42:53 -08:00
Matei Zaharia 7c3a1bddb7 Merge pull request #405 from JoshRosen/spark-530 
Add test for driver hanging on exit (SPARK-530)
 2013-01-22 23:28:47 -08:00
Tathagata Das bacade6caf Modified BlockManagerId API to ensure zero duplicate objects. Fixed BlockManagerId testcase in BlockManagerTestSuite. 2013-01-22 22:55:26 -08:00
Josh Rosen 43e9ff9596 Add test for driver hanging on exit (SPARK-530). 2013-01-22 22:47:26 -08:00
Charles Reiss 2849931000 Eliminate CacheTracker. 
Replaces DAGScheduler's queries of CacheTracker with BlockManagerMaster
queries.

Adds CacheManager to locally coordinate computation of cached RDDs.
 2013-01-22 22:19:30 -08:00
Matei Zaharia ebaa8f6519 Merge remote-tracking branch 'stephenh/cleanup' 
Conflicts:
	core/src/main/scala/spark/scheduler/local/LocalScheduler.scala
 2013-01-22 21:05:45 -08:00
Matei Zaharia d2d273868b Merge pull request #397 from JoshRosen/refactoring/daemon-threads 
Refactor daemon thread creation
 2013-01-22 21:02:53 -08:00
Matei Zaharia 05f13280de Merge pull request #404 from mbautin/master-fix-avro-dependency-in-repl 
Add an Avro dependency to REPL to make it compile with Hadoop 2
 2013-01-22 19:59:52 -08:00
Mikhail Bautin 325297e5c3 Add an Avro dependency to REPL to make it compile with Hadoop 2 2013-01-22 18:11:51 -08:00
Tathagata Das fad2b82fc8 Added support for saving input files of FileInputDStream to graph checkpoints. Modified 'file input stream with checkpoint' testcase to test recovery of pre-master-failure input files. 2013-01-22 18:10:00 -08:00
Josh Rosen 35168d9c89 Fix sys.path bug in PySpark SparkContext.addPyFile 2013-01-22 17:54:11 -08:00
Stephen Haberman 98d0b7747d Fix Worker logInfo about unknown executor. 2013-01-22 18:11:51 -06:00
Stephen Haberman 8c51322cd0 Don't bother creating an exception. 2013-01-22 18:09:10 -06:00
Stephen Haberman fdec42385a Fix SPARK_MEM in ExecutorRunner. 2013-01-22 18:01:12 -06:00
Stephen Haberman 2437f6741b Restore SPARK_MEM in executorEnvs. 2013-01-22 18:01:03 -06:00
Matei Zaharia 151c47eef5 Merge pull request #399 from NFLabs/master 
Fix for hanging spark.HttpFileServer on the kind of virtual network
 2013-01-22 15:49:24 -08:00
Matei Zaharia 49c78446c8 Merge pull request #402 from stephenh/handleslavelosttoosoon 
Handle slaveLost before slaveIdToHost knows about it.
 2013-01-22 15:48:37 -08:00
Stephen Haberman 250fe89679 Handle Master telling the Worker to kill an already-dead executor. 2013-01-22 16:29:05 -06:00
Stephen Haberman 6f2194f757 Call removeJob instead of killing the cluster. 2013-01-22 15:38:58 -06:00
Stephen Haberman 27b3f3f0a9 Handle slaveLost before slaveIdToHost knows about it. 2013-01-22 15:30:42 -06:00
Imran Rashid 905c720e5e Merge branch 'master' into blockmanager_ui 
Conflicts:
	core/src/main/scala/spark/RDD.scala
 2013-01-22 12:02:27 -08:00
Matei Zaharia d1de9d7d15 Merge pull request #400 from stephenh/sparkcontextcstrs 
Use default arguments instead of constructor overloads.
 2013-01-22 11:56:54 -08:00
Imran Rashid 50e2b23927 Fix up some problems from the merge 2013-01-22 11:46:01 -08:00
Stephen Haberman 588b24197a Use default arguments instead of constructor overloads. 2013-01-22 10:19:30 -06:00
Leemoonsoo 7e9ee2e833 Fix for hanging spark.HttpFileServer with kind of virtual network 2013-01-22 23:08:34 +09:00
Tathagata Das 364cdb679c Refactored DStreamCheckpointData. 2013-01-22 00:43:31 -08:00
Charles Reiss e353886a8c Use generation numbers for fetch failure tracking 2013-01-22 00:23:31 -08:00
Prashant Sharma d17065c4b5 actor as receiver 2013-01-22 13:28:29 +05:30
Josh Rosen 551a47a620 Refactor daemon thread pool creation. 2013-01-21 23:31:00 -08:00
Josh Rosen c75ae3622e Make AccumulatorParam an abstract base class. 2013-01-21 22:32:57 -08:00
Stephen Haberman a8baeb9327 Further simplify getOrElse call. 2013-01-21 21:30:24 -06:00
haitao.yao df9ae8a74e add ensime 2013-01-22 10:12:37 +08:00
Matei Zaharia 7d3e359f2c Merge pull request #395 from stephenh/fixrecursion 
Remove unneeded/now-broken saveAsNewAPIHadoopFile overload.
 2013-01-21 18:11:31 -08:00
Stephen Haberman 2d8218b871 Remove unneeded/now-broken saveAsNewAPIHadoopFile overload. 2013-01-21 20:00:27 -06:00
Josh Rosen 7b9e96c992 Add synchronization to Executor.updateDependencies() (SPARK-662) 2013-01-21 17:34:23 -08:00
Josh Rosen ef711902c1 Don't download files to master's working directory. 
This should avoid exceptions caused by existing
files with different contents.

I also removed some unused code.
 2013-01-21 17:34:17 -08:00