[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-12-01 Thread liyezhang556520
Github user liyezhang556520 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r21142893 --- Diff: core/src/main/scala/org/apache/spark/CacheManager.scala --- @@ -118,21 +118,29 @@ private[spark] class CacheManager(blockManager: BlockManag

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-09-16 Thread cloud-fan
Github user cloud-fan commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r17646960 --- Diff: core/src/main/scala/org/apache/spark/util/collection/SizeTrackingVector.scala --- @@ -0,0 +1,46 @@ +/* + * Licensed to the Apache Softwar

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-09-16 Thread cloud-fan
Github user cloud-fan commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-55850975 Hi @andrewor14 , I think about your fix and focus on the timing to release unrollMemory. If we unroll a partition successfully, currently we release unrollMemory immedi

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-09-16 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r17645466 --- Diff: core/src/main/scala/org/apache/spark/CacheManager.scala --- @@ -118,21 +118,29 @@ private[spark] class CacheManager(blockManager: BlockManager) e

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-09-16 Thread cloud-fan
Github user cloud-fan commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r17645275 --- Diff: core/src/main/scala/org/apache/spark/CacheManager.scala --- @@ -118,21 +118,29 @@ private[spark] class CacheManager(blockManager: BlockManager) ex

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-09-16 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r17617153 --- Diff: core/src/main/scala/org/apache/spark/CacheManager.scala --- @@ -118,21 +118,29 @@ private[spark] class CacheManager(blockManager: BlockManager) e

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-09-15 Thread cloud-fan
Github user cloud-fan commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r17584855 --- Diff: core/src/main/scala/org/apache/spark/CacheManager.scala --- @@ -118,21 +118,29 @@ private[spark] class CacheManager(blockManager: BlockManager) ex

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-09-15 Thread cloud-fan
Github user cloud-fan commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r17584609 --- Diff: core/src/main/scala/org/apache/spark/CacheManager.scala --- @@ -118,21 +118,29 @@ private[spark] class CacheManager(blockManager: BlockManager) ex

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-09-15 Thread cloud-fan
Github user cloud-fan commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r17581688 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +193,93 @@ private class MemoryStore(blockManager: BlockManager, maxM

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-09-15 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r17576857 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +193,93 @@ private class MemoryStore(blockManager: BlockManager, max

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-09-14 Thread cloud-fan
Github user cloud-fan commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r17518799 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +193,93 @@ private class MemoryStore(blockManager: BlockManager, maxM

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-09-12 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r17489658 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +193,93 @@ private class MemoryStore(blockManager: BlockManager, max

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-09-12 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r17489633 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +193,93 @@ private class MemoryStore(blockManager: BlockManager, max

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-09-12 Thread cloud-fan
Github user cloud-fan commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r17470235 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +193,93 @@ private class MemoryStore(blockManager: BlockManager, maxM

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-09-12 Thread cloud-fan
Github user cloud-fan commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r17469980 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +193,93 @@ private class MemoryStore(blockManager: BlockManager, maxM

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-08-26 Thread liyezhang556520
Github user liyezhang556520 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r16703247 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +193,93 @@ private class MemoryStore(blockManager: BlockManager,

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-08-25 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r16690496 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +193,93 @@ private class MemoryStore(blockManager: BlockManager, max

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-08-12 Thread liyezhang556520
Github user liyezhang556520 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r16100690 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +193,93 @@ private class MemoryStore(blockManager: BlockManager,

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-27 Thread andrewor14
Github user andrewor14 commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50290568 Great. Thanks for reviewing @pwendell @mateiz @mridulm. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well.

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-27 Thread asfgit
Github user asfgit closed the pull request at: https://github.com/apache/spark/pull/1165 --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enab

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-27 Thread mateiz
Github user mateiz commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50289038 Thanks Andrew! The changes look good to me -- I've merged this in. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub a

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-27 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50281752 QA results for PR 1165:- This patch PASSES unit tests.- This patch merges cleanly- This patch adds the following public classes (experimental):case class Sample(size: Lon

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-27 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50280496 QA tests have started for PR 1165. This patch merges cleanly. View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/17245/consoleFull --- If

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-27 Thread andrewor14
Github user andrewor14 commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50280386 I have addressed your latest comments and rebased to master. Anything else? --- If your project is set up for it, you can reply to this email and have your reply appea

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-26 Thread mateiz
Github user mateiz commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50246395 @andrewor14 I looked through this and it looks good to me. Made a few very small comments to clarify the algorithm. Thanks for adding the test with multiple blocks being u

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-26 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15435485 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +193,104 @@ private class MemoryStore(blockManager: BlockManager, maxMem

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-26 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15435457 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +193,104 @@ private class MemoryStore(blockManager: BlockManager, maxMem

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-26 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15435454 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +193,104 @@ private class MemoryStore(blockManager: BlockManager, maxMem

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-25 Thread andrewor14
Github user andrewor14 commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50219496 @mateiz There were a couple of correctness issues with the previous code that my latest commits have fixed (see commit message for more detail). After fixing these I h

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-25 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50219182 QA results for PR 1165:- This patch PASSES unit tests.- This patch merges cleanly- This patch adds the following public classes (experimental):case class Sample(size: Lon

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-25 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50217815 QA tests have started for PR 1165. This patch merges cleanly. View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/17208/consoleFull --- If

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-25 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15413324 --- Diff: core/src/test/scala/org/apache/spark/CacheManagerSuite.scala --- @@ -52,22 +50,21 @@ class CacheManagerSuite extends FunSuite with BeforeAndAfter

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-25 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15412910 --- Diff: core/src/test/scala/org/apache/spark/CacheManagerSuite.scala --- @@ -52,22 +50,21 @@ class CacheManagerSuite extends FunSuite with BeforeAndAfter

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-24 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50109743 QA results for PR 1165:- This patch PASSES unit tests.- This patch merges cleanly- This patch adds the following public classes (experimental):case class Sample(size: Lon

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-24 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50107921 QA tests have started for PR 1165. This patch merges cleanly. View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/17162/consoleFull --- If

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-24 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50102333 QA results for PR 1165:- This patch FAILED unit tests.- This patch merges cleanly- This patch adds the following public classes (experimental):case class Sample(size: Lon

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-24 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15383252 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -275,8 +426,36 @@ private class MemoryStore(blockManager: BlockManager, max

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-24 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15383242 --- Diff: core/src/test/scala/org/apache/spark/CacheManagerSuite.scala --- @@ -52,22 +50,21 @@ class CacheManagerSuite extends FunSuite with BeforeAndAfter wit

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-24 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15383171 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -275,8 +426,36 @@ private class MemoryStore(blockManager: BlockManager, maxMemo

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-24 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15383154 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -275,8 +426,36 @@ private class MemoryStore(blockManager: BlockManager, maxMemo

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-24 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15383132 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -20,25 +20,43 @@ package org.apache.spark.storage import java.nio.ByteBuffe

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-24 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50099588 QA tests have started for PR 1165. This patch merges cleanly. View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/17155/consoleFull --- If

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-24 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50097869 QA results for PR 1165:- This patch FAILED unit tests.- This patch merges cleanly- This patch adds the following public classes (experimental):case class Sample(size: Lon

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-24 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50096538 QA tests have started for PR 1165. This patch merges cleanly. View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/17144/consoleFull --- If

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-24 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50074496 QA results for PR 1165:- This patch FAILED unit tests.- This patch merges cleanly- This patch adds the following public classes (experimental):case class Sample(size: Lon

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-24 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-50069722 QA tests have started for PR 1165. This patch merges cleanly. View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/17133/consoleFull --- If

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-24 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15363550 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -215,7 +361,10 @@ private class MemoryStore(blockManager: BlockManager, max

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-23 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15331347 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -87,9 +97,47 @@ private class MemoryStore(blockManager: BlockManager, maxMemory

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-23 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15331245 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +189,104 @@ private class MemoryStore(blockManager: BlockManager, maxMem

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-23 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15331224 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +189,104 @@ private class MemoryStore(blockManager: BlockManager, maxMem

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-23 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49970876 QA results for PR 1165:- This patch FAILED unit tests.- This patch merges cleanly- This patch adds the following public classes (experimental):case class Sample(size: Lon

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-23 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15331171 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +189,104 @@ private class MemoryStore(blockManager: BlockManager, maxMem

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-23 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15331108 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -215,7 +361,10 @@ private class MemoryStore(blockManager: BlockManager, maxMemo

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-23 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15331070 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -215,7 +361,10 @@ private class MemoryStore(blockManager: BlockManager, maxMemo

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-23 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15331051 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -225,9 +374,18 @@ private class MemoryStore(blockManager: BlockManager, maxMemo

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-23 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15331032 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +189,104 @@ private class MemoryStore(blockManager: BlockManager, maxMem

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-23 Thread andrewor14
Github user andrewor14 commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49969030 @pwendell @mateiz I have updated the unroll logic to do something similar to EAOM. More details are in the commit message. I have also added some synchronization logic

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-23 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49968995 QA tests have started for PR 1165. This patch merges cleanly. View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/17099/consoleFull --- If

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread pwendell
Github user pwendell commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15271976 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +188,88 @@ private class MemoryStore(blockManager: BlockManager, maxMe

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread mateiz
Github user mateiz commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49831796 BTW for these changes it would be good to perf-test it, e.g. design a workload where your blocks do fit in memory (just on one multicore machine is fine) and make sure the

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15270521 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -87,9 +97,46 @@ private class MemoryStore(blockManager: BlockManager, maxMemory

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15270516 --- Diff: core/src/main/scala/org/apache/spark/CacheManager.scala --- @@ -140,14 +148,36 @@ private[spark] class CacheManager(blockManager: BlockManager) exten

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread mateiz
Github user mateiz commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49831426 Hey @andrewor14 I had some comments on the way this decides when to check memory usage. It seems quite expensive in the current form. --- If your project is set up for it

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15270480 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +188,88 @@ private class MemoryStore(blockManager: BlockManager, maxMemo

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15270415 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +188,88 @@ private class MemoryStore(blockManager: BlockManager, maxMemo

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15270377 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +188,88 @@ private class MemoryStore(blockManager: BlockManager, maxMemo

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15270358 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +188,88 @@ private class MemoryStore(blockManager: BlockManager, maxMemo

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49809451 QA results for PR 1165:- This patch PASSES unit tests.- This patch merges cleanly- This patch adds the following public classes (experimental):case class Sample(size: Lon

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49797911 QA tests have started for PR 1165. This patch merges cleanly. View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/16986/consoleFull --- If

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread andrewor14
Github user andrewor14 commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49797626 test this please! --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this fe

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread andrewor14
Github user andrewor14 commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49797163 test this please --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this fea

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread andrewor14
Github user andrewor14 commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49797224 test this please --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this fea

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49795173 QA results for PR 1165:- This patch FAILED unit tests.- This patch merges cleanly- This patch adds the following public classes (experimental):case class Sample(size: Lon

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49782460 QA tests have started for PR 1165. This patch merges cleanly. View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/16979/consoleFull --- If

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15242324 --- Diff: core/src/main/scala/org/apache/spark/CacheManager.scala --- @@ -140,14 +145,36 @@ private[spark] class CacheManager(blockManager: BlockManager) e

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread pwendell
Github user pwendell commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15214991 --- Diff: core/src/main/scala/org/apache/spark/util/collection/SizeTracker.scala --- @@ -0,0 +1,105 @@ +/* + * Licensed to the Apache Software Founda

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread pwendell
Github user pwendell commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15214933 --- Diff: core/src/main/scala/org/apache/spark/CacheManager.scala --- @@ -140,14 +145,36 @@ private[spark] class CacheManager(blockManager: BlockManager) ext

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread pwendell
Github user pwendell commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15214812 --- Diff: core/src/main/scala/org/apache/spark/CacheManager.scala --- @@ -124,15 +124,20 @@ private[spark] class CacheManager(blockManager: BlockManager) ext

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-22 Thread pwendell
Github user pwendell commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15214670 --- Diff: core/src/main/scala/org/apache/spark/CacheManager.scala --- @@ -124,15 +124,20 @@ private[spark] class CacheManager(blockManager: BlockManager) ext

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-21 Thread mateiz
Github user mateiz commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49685834 I see, we should revisit that, but probably later. IMO if you're going to store them serialized, you're going to pay for that deserialization cost later anyway. With the c

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-21 Thread andrewor14
Github user andrewor14 commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49685219 @mateiz Yes, currently we unroll it in deserialized form even if we only want to store it in serialized form. One issue with storing it directly in bytes is that `Cach

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-21 Thread mateiz
Github user mateiz commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49682053 Hey @andrewor14, one question here just to make sure I understand: if the data is supposed to be stored as MEMORY_ONLY_SER, will this code still unroll it in an un-seriali

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-21 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15195362 --- Diff: core/src/main/scala/org/apache/spark/storage/BlockManager.scala --- @@ -561,13 +562,14 @@ private[spark] class BlockManager( iter

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-21 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15195205 --- Diff: core/src/main/scala/org/apache/spark/storage/BlockManager.scala --- @@ -463,16 +463,15 @@ private[spark] class BlockManager( v

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-21 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15194738 --- Diff: core/src/main/scala/org/apache/spark/util/collection/SizeTrackingAppendOnlyBuffer.scala --- @@ -0,0 +1,43 @@ +/* + * Licensed to the Apa

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-21 Thread andrewor14
Github user andrewor14 commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15194471 --- Diff: core/src/main/scala/org/apache/spark/CacheManager.scala --- @@ -140,14 +145,36 @@ private[spark] class CacheManager(blockManager: BlockManager) e

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-20 Thread colorant
Github user colorant commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15153652 --- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala --- @@ -141,6 +188,88 @@ private class MemoryStore(blockManager: BlockManager, maxMe

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-19 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15147430 --- Diff: core/src/main/scala/org/apache/spark/util/collection/SizeTracker.scala --- @@ -0,0 +1,105 @@ +/* + * Licensed to the Apache Software Foundati

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-19 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15147427 --- Diff: core/src/main/scala/org/apache/spark/CacheManager.scala --- @@ -140,14 +145,36 @@ private[spark] class CacheManager(blockManager: BlockManager) exten

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-19 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15147423 --- Diff: core/src/main/scala/org/apache/spark/storage/BlockManager.scala --- @@ -463,16 +463,15 @@ private[spark] class BlockManager( val v

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-19 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15147419 --- Diff: core/src/main/scala/org/apache/spark/storage/BlockManager.scala --- @@ -463,16 +463,15 @@ private[spark] class BlockManager( val v

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-19 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15147416 --- Diff: core/src/main/scala/org/apache/spark/util/collection/SizeTrackingAppendOnlyBuffer.scala --- @@ -0,0 +1,43 @@ +/* + * Licensed to the Apache

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-19 Thread mateiz
Github user mateiz commented on a diff in the pull request: https://github.com/apache/spark/pull/1165#discussion_r15145476 --- Diff: core/src/main/scala/org/apache/spark/util/collection/SizeTrackingAppendOnlyBuffer.scala --- @@ -0,0 +1,43 @@ +/* + * Licensed to the Apache

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-19 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49516746 QA tests have started for PR 1165. This patch merges cleanly. View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/16851/consoleFull --- If

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-18 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49497172 QA results for PR 1165:- This patch PASSES unit tests.- This patch merges cleanly- This patch adds the following public classes (experimental):case class Sample(size: Lon

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-18 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49494682 QA tests have started for PR 1165. This patch merges cleanly. View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/16840/consoleFull --- If

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-18 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49491986 QA results for PR 1165:- This patch PASSES unit tests.- This patch merges cleanly- This patch adds the following public classes (experimental):case class Sample(size: Lon

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-18 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49485183 QA tests have started for PR 1165. This patch merges cleanly. View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/16836/consoleFull --- If

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-18 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49475344 QA results for PR 1165:- This patch FAILED unit tests.- This patch merges cleanly- This patch adds the following public classes (experimental):case class Sample(size: Lon

[GitHub] spark pull request: [SPARK-1777] Prevent OOMs from single partitio...

2014-07-18 Thread SparkQA
Github user SparkQA commented on the pull request: https://github.com/apache/spark/pull/1165#issuecomment-49465699 QA tests have started for PR 1165. This patch merges cleanly. View progress: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/16828/consoleFull --- If

  1   2   >