Commit graph

12 commits

Author SHA1 Message Date
liguoqiang b662c88a24 fix this import order 2013-12-26 15:49:33 +08:00
liguoqiang 2bd76f693d Renamed ClusterScheduler to TaskSchedulerImpl for yarn and new-yarn 2013-12-26 11:10:35 +08:00
liguoqiang 14fcef72db Renamed ClusterScheduler to TaskSchedulerImpl for yarn and new-yarn 2013-12-26 11:05:07 +08:00
azuryyu 66b7bea7f8 Make App report interval configurable during 'run on Yarn' 2013-12-24 18:16:49 +08:00
Patrick Wendell 10c0ffa1eb One other fix 2013-12-16 22:10:55 -08:00
Patrick Wendell c1c0f8099f Clean-up 2013-12-16 22:01:27 -08:00
Patrick Wendell 24f8220dc8 Removing extra code in new yarn 2013-12-16 21:53:51 -08:00
Patrick Wendell c6f95e603e Attempt with extra repositories 2013-12-16 21:53:51 -08:00
Thomas Graves 842eb55fb5 Fix the --name option for Spark on Yarn 2013-12-12 11:11:09 -06:00
Raymond Liu 403cac9be3 Port yarn-client mode for new-yarn 2013-11-27 16:10:42 +08:00
Harvey Feng 4f1c3fa5d7 Hadoop 2.2 YARN API migration for SPARK_HOME/new-yarn 2013-11-23 17:08:30 -08:00
Harvey Feng ab8652f2d3 Add a "new-yarn" directory in SPARK_HOME, intended to contain Hadoop-2.2 API changes. 2013-11-23 17:08:30 -08:00