379567f | jerryshao | 30 May 2013, 09:55:11 UTC | fix block manager UI display issue when enable spark.cleaner.ttl Conflicts: core/src/main/scala/spark/storage/StorageUtils.scala | 22 June 2013, 15:51:01 UTC |
2462b40 | Christopher Nguyen | 05 June 2013, 05:12:47 UTC | In the current code, when both partitions happen to have zero-length, the return mean will be NaN. Consequently, the result of mean after reducing over all partitions will also be NaN, which is not correct if there are partitions with non-zero length. This patch fixes this issue. | 22 June 2013, 15:40:12 UTC |
5539549 | Jey Kottalam | 21 June 2013, 16:13:48 UTC | Fix reporting of PySpark exceptions | 22 June 2013, 14:58:36 UTC |
439f2b4 | Jey Kottalam | 20 June 2013, 16:49:10 UTC | use parens when calling method with side-effects | 22 June 2013, 14:58:32 UTC |
e45450c | Jey Kottalam | 23 May 2013, 18:50:24 UTC | PySpark daemon: fix deadlock, improve error handling | 22 June 2013, 14:58:27 UTC |
104389b | Jey Kottalam | 13 May 2013, 15:53:47 UTC | Rename PythonWorker to PythonWorkerFactory | 22 June 2013, 14:58:13 UTC |
126f134 | Jey Kottalam | 10 May 2013, 22:48:48 UTC | Add tests and fixes for Python daemon shutdown | 22 June 2013, 14:58:08 UTC |
c294fa8 | Jey Kottalam | 06 May 2013, 23:34:30 UTC | Prefork Python worker processes Conflicts: core/src/main/scala/spark/SparkEnv.scala | 22 June 2013, 14:57:53 UTC |
13faa48 | Jey Kottalam | 10 March 2013, 20:54:46 UTC | Add Python timing instrumentation | 22 June 2013, 14:56:40 UTC |
40b92ce | Reynold Xin | 05 June 2013, 05:35:42 UTC | Merge pull request #642 from pwendell/branch-0.7 Bump akka and blockmanager timeouts to 60 seconds (branch 0.7 backport) | 05 June 2013, 05:35:42 UTC |
9085ebf | Patrick Wendell | 04 June 2013, 22:14:24 UTC | Bump akka and blockmanager timeouts to 60 seconds | 04 June 2013, 23:14:38 UTC |
bf2e0c6 | Josh Rosen | 05 May 2013, 18:30:36 UTC | Fix SPARK-670: EC2 start command should require -i option. | 03 June 2013, 00:27:21 UTC |
e27c058 | Josh Rosen | 16 May 2013, 08:52:40 UTC | Abort job if result exceeds Akka frame size; add test. Conflicts: core/src/test/scala/spark/DistributedSuite.scala | 03 June 2013, 00:27:09 UTC |
d396c42 | Josh Rosen | 12 May 2013, 01:27:26 UTC | Throw exception if task result exceeds Akka frame size. This partially addresses SPARK-747. | 03 June 2013, 00:16:32 UTC |
f29d883 | jerryshao | 24 May 2013, 06:26:19 UTC | fix CheckpointRDD getPreferredLocations java.io.FileNotFoundException | 03 June 2013, 00:14:37 UTC |
d77e73c | Matei Zaharia | 02 June 2013, 23:20:08 UTC | Merge pull request #609 from pwendell/SPARK-738-0.7 SPARK-738 (branch 0.7 backport) Fixing nonserializable exception bug | 02 June 2013, 23:20:08 UTC |
f3891f1 | Matei Zaharia | 02 June 2013, 23:17:33 UTC | Use correct API directory in doc generation | 02 June 2013, 23:17:33 UTC |
eb19b9a | Jey Kottalam | 29 May 2013, 04:05:21 UTC | Update version to 0.7.3-SNAPSHOT | 02 June 2013, 23:17:26 UTC |
86cc03b | Matei Zaharia | 02 June 2013, 00:12:22 UTC | Revert "Update version to 0.7.3-SNAPSHOT" This reverts commit e5fbdac22acba28d6691901a2a9afcf37a80cc74. | 02 June 2013, 00:12:22 UTC |
c0e1ad5 | Matei Zaharia | 01 June 2013, 23:42:17 UTC | Merge remote-tracking branch 'jey/branch-0.7' into branch-0.7 | 01 June 2013, 23:42:17 UTC |
e5fbdac | Jey Kottalam | 29 May 2013, 04:05:21 UTC | Update version to 0.7.3-SNAPSHOT | 29 May 2013, 04:05:21 UTC |
d8fb7b6 | Jey Kottalam | 28 May 2013, 22:32:10 UTC | Update latest AMI URL for 0.7.2 | 29 May 2013, 03:51:13 UTC |
d61956f | Jey Kottalam | 25 May 2013, 01:03:07 UTC | Include URI scheme when generating cluster_url | 29 May 2013, 03:51:13 UTC |
2d84917 | Josh Rosen | 24 May 2013, 20:05:06 UTC | Use ec2-metadata in start-slave.sh. PR #419 applied the same change, but only to start-master.sh, so some workers were still starting their web UI's using internal addresses. This should finally fix SPARK-613. | 26 May 2013, 06:38:59 UTC |
587c1f0 | Jey Kottalam | 18 May 2013, 23:11:56 UTC | Release v0.7.2 | 18 May 2013, 23:11:56 UTC |
ca398f2 | Jey Kottalam | 18 May 2013, 23:11:29 UTC | Exclude old versions of Netty from Maven-based build | 18 May 2013, 23:11:42 UTC |
2e134e9 | Matei Zaharia | 26 April 2013, 02:52:12 UTC | Exclude old versions of Netty, which had a different Maven organization | 18 May 2013, 23:11:42 UTC |
c2c3597 | Jey Kottalam | 17 May 2013, 18:59:55 UTC | repl shouldn't depend on bagel and examples | 17 May 2013, 19:00:02 UTC |
1f18db0 | Jey Kottalam | 17 May 2013, 03:22:33 UTC | Fix generation of *-javadoc.jar files | 17 May 2013, 04:18:55 UTC |
9632bc1 | Jey Kottalam | 07 May 2013, 23:58:11 UTC | Update run script to find updated JARs | 17 May 2013, 04:18:55 UTC |
ab4b3a2 | Jey Kottalam | 15 May 2013, 22:42:18 UTC | Fix Hadoop2 build. Enable with -Dhadoop.version=2 | 17 May 2013, 04:18:55 UTC |
2a9a44c | Jey Kottalam | 15 May 2013, 17:53:27 UTC | Build Hadoop1 variant of Spark by default | 17 May 2013, 04:18:55 UTC |
4fb04bb | Jey Kottalam | 14 May 2013, 22:30:31 UTC | Make Maven artifactIds constant | 16 May 2013, 21:27:50 UTC |
20c4290 | Jey Kottalam | 16 May 2013, 21:27:35 UTC | Merge remote-tracking branch 'origin/branch-0.7' into branch-0.7 | 16 May 2013, 21:27:35 UTC |
9b77a6f | Patrick Wendell | 12 May 2013, 14:54:03 UTC | Capturing class name | 12 May 2013, 18:49:28 UTC |
e80fe37 | Patrick Wendell | 12 May 2013, 06:53:50 UTC | Small fix | 12 May 2013, 18:48:42 UTC |
507dc75 | Patrick Wendell | 12 May 2013, 06:52:53 UTC | Removing import | 12 May 2013, 18:48:32 UTC |
51a6fbc | Patrick Wendell | 12 May 2013, 06:39:14 UTC | Changing technique to use same code path in all cases | 12 May 2013, 18:48:22 UTC |
2e560b7 | Patrick Wendell | 11 May 2013, 21:20:39 UTC | Removing unnecessary map | 12 May 2013, 18:48:16 UTC |
24d897b | Patrick Wendell | 11 May 2013, 21:17:09 UTC | SPARK-738: Spark should detect and squash nonserializable exceptions | 12 May 2013, 18:48:06 UTC |
14a88a1 | Mark Hamstra | 10 May 2013, 20:02:34 UTC | Actually use the cleaned closure in foreachPartition | 10 May 2013, 20:57:10 UTC |
8115127 | Reynold Xin | 30 April 2013, 23:33:56 UTC | Merge pull request #588 from pwendell/branch-0.7 Spark-742: TaskMetrics should not employ per-record timing. | 30 April 2013, 23:33:56 UTC |
cb027f7 | Patrick Wendell | 29 April 2013, 17:10:14 UTC | Spark-742: TaskMetrics should not employ per-record timing. This commit merges together all the changes for purpose of cherry-picking | 30 April 2013, 22:16:33 UTC |
15dbbd0 | Jey Kottalam | 25 April 2013, 23:38:56 UTC | Update development version to 0.7.2-SNAPSHOT | 26 April 2013, 22:50:04 UTC |
00e78b6 | Jey Kottalam | 26 April 2013, 17:58:40 UTC | Release v0.7.1 | 26 April 2013, 17:58:40 UTC |
9d06bc9 | Jey Kottalam | 26 April 2013, 05:54:47 UTC | Mark kafka:0.7.2-spark as an externally provided dependency | 26 April 2013, 17:58:18 UTC |
20b0eb1 | Jey Kottalam | 25 April 2013, 21:01:15 UTC | Include scala version in artifactId | 25 April 2013, 22:53:15 UTC |
95a9924 | Jey Kottalam | 24 April 2013, 00:44:14 UTC | Conform to Sonatype's "Central Sync Requirements" | 25 April 2013, 09:04:01 UTC |
ee93c18 | Jey Kottalam | 23 April 2013, 03:04:10 UTC | Configure Maven to perform releases | 25 April 2013, 09:03:57 UTC |
2e97618 | Matei Zaharia | 25 April 2013, 01:53:12 UTC | Create an empty directory when checkpointing a 0-partition RDD (fixes a test failure on Hadoop 2.0) | 25 April 2013, 01:53:12 UTC |
f216ea0 | Matei Zaharia | 24 April 2013, 22:51:59 UTC | Call commitJob on Hadoop OutputFormats to fix issues with saving on Hadoop 2 | 24 April 2013, 22:51:59 UTC |
a9b3856 | Matei Zaharia | 24 April 2013, 22:08:20 UTC | Update unit test memory to 2 GB | 24 April 2013, 22:08:20 UTC |
80efd7c | Matei Zaharia | 20 April 2013, 00:14:36 UTC | Merge pull request #574 from jey/make-repl-not-require-streaming Remove spark-repl's extraneous dependency on spark-streaming | 20 April 2013, 00:14:36 UTC |
b9810ea | Matei Zaharia | 19 April 2013, 06:41:32 UTC | Merge pull request #576 from jey/update-maven-to-scala-2.9.3 Update Maven build to Scala 2.9.3 | 19 April 2013, 06:41:32 UTC |
f5756cd | Matei Zaharia | 19 April 2013, 05:25:21 UTC | Turn on forking in test JVMs to reduce the pressure on perm gen and code cache sizes due to having 2 instances of the Scala compiler and a bunch of classloaders. | 19 April 2013, 05:25:21 UTC |
c9b1811 | Jey Kottalam | 18 April 2013, 21:58:38 UTC | Update Maven build to Scala 2.9.3 | 18 April 2013, 21:58:38 UTC |
b9aef50 | Jey Kottalam | 18 April 2013, 19:08:11 UTC | Remove spark-repl's extraneous dependency on spark-streaming | 18 April 2013, 19:08:11 UTC |
7e1ec8c | Jey Kottalam | 16 April 2013, 06:32:25 UTC | Increase ReservedCodeCacheSize for sbt | 17 April 2013, 02:48:41 UTC |
56e3ca0 | Matei Zaharia | 15 April 2013, 20:41:51 UTC | Make ShuffledRDD.prev transient | 15 April 2013, 20:42:16 UTC |
1c2bb35 | Andrew Ash | 09 April 2013, 18:35:29 UTC | Don't pull in old versions of Jackson via hadoop-core | 14 April 2013, 15:26:07 UTC |
c3769c7 | Mike | 12 April 2013, 03:52:06 UTC | Reversed the order of tests to find a scala executable (in the case when SPARK_LAUNCH_WITH_SCALA is defined): instead of checking in the PATH first, and only then (if not found) for SCALA_HOME, now we check for SCALA_HOME first, and only then (if not defined) do we look in the PATH. The advantage is that now if the user has a more recent (non-compatible) version of scala in her PATH, she can use SCALA_HOME to point to the older (compatible) version for use with spark. Suggested by Josh Rosen in this thread: https://groups.google.com/forum/?fromgroups=#!topic/spark-users/NC9JKvP8808 | 14 April 2013, 15:13:28 UTC |
ceb3c4c | Matei Zaharia | 12 April 2013, 02:29:37 UTC | Fix tests from different projects running in parallel in SBT 0.12 | 12 April 2013, 02:30:24 UTC |
f8ca144 | Andrew Ash | 10 April 2013, 03:35:13 UTC | Uniform whitespace across scala examples | 10 April 2013, 05:07:53 UTC |
c2f537b | Holden Karau | 10 April 2013, 04:37:02 UTC | Just use a loop for retries | 10 April 2013, 05:05:37 UTC |
c9baafc | Holden Karau | 06 April 2013, 07:35:50 UTC | Retry failed ssh commands. This is especially useful during system startup when the hosts may not have yet come on-line but can be useful at other points for people with flakey connections | 10 April 2013, 05:05:37 UTC |
ef42c30 | Matei Zaharia | 08 April 2013, 21:34:38 UTC | Fix passing of superstep in Bagel to avoid seeing new values of the superstep value upon recomputation, and set the default storage level in Bagel to MEMORY_AND_DISK | 08 April 2013, 21:35:21 UTC |
c01958f | Patrick Wendell | 08 April 2013, 00:44:48 UTC | Updating based on code review | 08 April 2013, 16:45:24 UTC |
f232223 | Patrick Wendell | 03 April 2013, 21:29:46 UTC | SPARK-724 - Have Akka logging enabled by default for standalone daemons See the JIRA for more details. I was only able to test the bash version (don't have Windows) so maybe check over that the syntax is correct there. | 08 April 2013, 16:45:24 UTC |
054feb6 | Matei Zaharia | 08 April 2013, 01:15:21 UTC | Fixed a bug with zip | 08 April 2013, 01:15:21 UTC |
b5900d4 | Matei Zaharia | 08 April 2013, 00:55:42 UTC | Fix compile warning | 08 April 2013, 00:55:42 UTC |
1cb3eb9 | Matei Zaharia | 08 April 2013, 00:54:18 UTC | Merge remote-tracking branch 'kalpit/master' Conflicts: project/SparkBuild.scala | 08 April 2013, 00:54:18 UTC |
a158641 | Matei Zaharia | 08 April 2013, 00:31:19 UTC | Updated link to SBT | 08 April 2013, 00:31:19 UTC |
6962d40 | Matei Zaharia | 08 April 2013, 00:27:33 UTC | Fix deprecated warning | 08 April 2013, 00:27:33 UTC |
34a47b8 | Matei Zaharia | 08 April 2013, 00:27:03 UTC | Update Scala version in docs | 08 April 2013, 00:27:03 UTC |
b362df3 | Matei Zaharia | 08 April 2013, 00:17:52 UTC | Merge pull request #552 from MLnick/master Bumping version for Twitter Algebird to latest | 08 April 2013, 00:17:52 UTC |
4b30190 | Matei Zaharia | 08 April 2013, 00:15:10 UTC | Merge pull request #554 from andyk/scala2.9.3 Fixes SPARK-723 - Update build to Scala 2.9.3 | 08 April 2013, 00:15:10 UTC |
5555811 | Andy Konwinski | 04 April 2013, 20:26:45 UTC | Update build to Scala 2.9.3 | 04 April 2013, 20:26:45 UTC |
0f54344 | Nick Pentreath | 03 April 2013, 11:15:34 UTC | Bumping Algebird version in examples now that it supports JDK 1.6 | 03 April 2013, 11:15:34 UTC |
dfe98ca | Matei Zaharia | 03 April 2013, 02:24:12 UTC | Merge pull request #550 from erikvanoosten/master corrected Algebird example | 03 April 2013, 02:24:12 UTC |
b5d7830 | Matei Zaharia | 03 April 2013, 02:23:45 UTC | Merge pull request #551 from jey/python-bugfixes Python bugfixes | 03 April 2013, 02:23:45 UTC |
9a731f5 | Jey Kottalam | 02 April 2013, 18:58:40 UTC | Fix Python saveAsTextFile doctest to not expect order to be preserved | 02 April 2013, 18:59:20 UTC |
2060400 | Jey Kottalam | 10 March 2013, 06:16:19 UTC | Fix argv handling in Python transitive closure example | 02 April 2013, 18:59:07 UTC |
b5e60c3 | Erik van oosten | 02 April 2013, 12:25:22 UTC | Corrected order of CountMinSketchMonoid arguments | 02 April 2013, 12:25:22 UTC |
2be2295 | Matei Zaharia | 01 April 2013, 01:09:14 UTC | Merge pull request #548 from markhamstra/getWritableClass_filter Fixed broken filter in getWritableClass[T] | 01 April 2013, 01:09:14 UTC |
e215f67 | Mark Hamstra | 31 March 2013, 15:00:13 UTC | Correct sense of 'filter out' in comment. | 31 March 2013, 15:00:13 UTC |
8bcdc64 | Mark Hamstra | 31 March 2013, 05:09:52 UTC | Fixed broken filter in getWritableClass[T] | 31 March 2013, 05:09:52 UTC |
9831bc1 | Matei Zaharia | 30 March 2013, 05:16:22 UTC | Merge pull request #539 from cgrothaus/fix-webui-workdirpath Bugfix: WorkerWebUI must respect workDirPath from Worker | 30 March 2013, 05:16:22 UTC |
3cc8ab6 | Matei Zaharia | 30 March 2013, 05:14:07 UTC | Merge pull request #541 from stephenh/shufflecoalesce Add a shuffle parameter to coalesce. | 30 March 2013, 05:14:07 UTC |
cad507a | Matei Zaharia | 30 March 2013, 05:13:12 UTC | Merge pull request #547 from jey/maven-streaming-tests-initialization-fix Move streaming test initialization into 'before' blocks | 30 March 2013, 05:13:12 UTC |
a98996d | Matei Zaharia | 30 March 2013, 05:12:15 UTC | Merge pull request #545 from ash211/patch-1 Don't use deprecated Application in example | 30 March 2013, 05:12:15 UTC |
104c694 | Matei Zaharia | 30 March 2013, 05:11:50 UTC | Merge pull request #546 from ash211/patch-2 Update tuning.md | 30 March 2013, 05:11:50 UTC |
e8f3669 | Andrew Ash | 28 March 2013, 22:17:39 UTC | Update tuning.md Make the example more compilable | 28 March 2013, 22:17:39 UTC |
b569b3f | Jey Kottalam | 27 March 2013, 01:14:04 UTC | Move streaming test initialization into 'before' blocks | 28 March 2013, 22:08:41 UTC |
4e2c965 | Andrew Ash | 28 March 2013, 20:47:37 UTC | Don't use deprecated Application in example As of 2.9.0 extending from Application is not recommended http://www.scala-lang.org/api/2.9.3/index.html#scala.Application | 28 March 2013, 20:47:37 UTC |
aa9134f | kalpit | 09 March 2013, 02:28:43 UTC | spark instance number must be present in log filename to prevent multiple workers from overriding each other's logs | 27 March 2013, 00:49:30 UTC |
f08db01 | kalpit | 07 March 2013, 02:06:32 UTC | added SPARK_WORKER_INSTANCES : allows spawning multiple worker instances/processes on every slave machine | 27 March 2013, 00:49:30 UTC |
f0164e5 | kalpit | 07 March 2013, 01:03:15 UTC | upgraded sbt version, sbt plugins and some library dependencies to latest stable version | 27 March 2013, 00:49:29 UTC |
bc36ee4 | Matei Zaharia | 26 March 2013, 22:05:13 UTC | Merge pull request #543 from holdenk/master Re-enable deprecation warnings and fix deprecated warning. | 26 March 2013, 22:05:13 UTC |
6c18131 | Andy Konwinski | 26 March 2013, 21:28:08 UTC | Add comment to README that 2.10 not yet supported | 26 March 2013, 21:28:08 UTC |
1f53811 | Holden Karau | 24 March 2013, 23:21:51 UTC | method first in trait IterableLike is deprecated: use `head' instead | 25 March 2013, 02:19:40 UTC |