Github user witgo closed the pull request at:
https://github.com/apache/spark/pull/11986
---
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
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/12113#discussion_r58324785
--- Diff: core/src/main/scala/org/apache/spark/MapOutputTracker.scala ---
@@ -428,40 +503,93 @@ private[spark] class MapOutputTrackerMaster(conf:
SparkConf
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/12113#discussion_r58325719
--- Diff: core/src/main/scala/org/apache/spark/MapOutputTracker.scala ---
@@ -296,10 +290,89 @@ private[spark] class MapOutputTrackerMaster(conf:
SparkConf
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/12113#discussion_r58502169
--- Diff: core/src/main/scala/org/apache/spark/MapOutputTracker.scala ---
@@ -296,10 +290,89 @@ private[spark] class MapOutputTrackerMaster(conf:
SparkConf
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/12113#discussion_r58503061
--- Diff: core/src/main/scala/org/apache/spark/MapOutputTracker.scala ---
@@ -428,40 +503,93 @@ private[spark] class MapOutputTrackerMaster(conf:
SparkConf
Github user witgo commented on the pull request:
https://github.com/apache/spark/pull/11986#issuecomment-202778045
retest 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 feature
Github user witgo commented on the pull request:
https://github.com/apache/spark/pull/12423#issuecomment-211219477
I think the following should be deleted
1.
[FileShuffleBlockResolver.scala](https://github.com/apache/spark/blob/master/core/src/main/scala/org/apache/spark/shuffle
GitHub user witgo opened a pull request:
https://github.com/apache/spark/pull/14647
[WIP][Test only][SPARK-6235]Address various 2G limits
## What changes were proposed in this pull request?
The main changes.
1. Replace DiskStore method `def getBytes (blockId: BlockId
GitHub user witgo opened a pull request:
https://github.com/apache/spark/pull/14662
[WIP][SPARK-17082][CORE]Replace ByteBuffer with ChunkedByteBuffer
## What changes were proposed in this pull request?
Replace `ByteBuffer` with `ChunkedByteBuffer
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/14647
@hvanhovell
I will submit some small PRs and provide a more high level description of
them.
---
If your project is set up for it, you can reply to this email and have your
reply appear
GitHub user witgo opened a pull request:
https://github.com/apache/spark/pull/14658
[WIP][SPARK-5928] Remote Shuffle Blocks cannot be more than 2 GB
## What changes were proposed in this pull request?
Add class `ChunkFetchInputStream` and it have the following effects
GitHub user witgo opened a pull request:
https://github.com/apache/spark/pull/14664
Spark 6236 caching blocks 2G
## What changes were proposed in this pull request?
(Please fill in changes proposed in this fix)
## How was this patch tested?
(Please
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/14647
@hvanhovell #14658 , #14662 and #14664
---
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
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r99458252
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskDescription.scala ---
@@ -51,8 +54,39 @@ private[spark] class TaskDescription(
val index
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r99457658
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -243,27 +245,42 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r99458022
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskDescription.scala ---
@@ -66,7 +100,8 @@ private[spark] object TaskDescription
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r99458165
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskDescription.scala ---
@@ -51,8 +54,39 @@ private[spark] class TaskDescription(
val index
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r99848516
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -243,27 +245,42 @@ class
Github user witgo closed the pull request at:
https://github.com/apache/spark/pull/16806
---
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
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/15505
Okay, this may take some time.
---
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
GitHub user witgo opened a pull request:
https://github.com/apache/spark/pull/16806
[WIP][SPARK-18890][CORE] Move task serialization from the TaskSetManager to
the CoarseGrainedSchedulerBackend
## What changes were proposed in this pull request?
See https
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r99462808
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -243,27 +245,42 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r99462431
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskDescription.scala ---
@@ -66,7 +100,8 @@ private[spark] object TaskDescription
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r99462387
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -243,27 +245,42 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r99463382
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -244,32 +245,45 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r97211797
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -602,6 +619,20 @@ class
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/15505
@squito
My understanding is that the TaskSchedulerImpl class contains many
synchronized statements (synchronized the methods). If a synchronized
statements execution time is very long
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r97695455
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -602,6 +619,21 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/16677#discussion_r102891969
--- Diff: core/src/main/scala/org/apache/spark/scheduler/MapStatus.scala ---
@@ -39,16 +40,18 @@ private[spark] sealed trait MapStatus {
* necessary
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/15505
@kayousterhout I think the latest code is ready to merge into the master
branch.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well
GitHub user witgo opened a pull request:
https://github.com/apache/spark/pull/14751
[WIP][SPARK-17184][[CORE]]Replace ByteBuf with InputStream
## What changes were proposed in this pull request?
The size of ByteBuf can not be greater than 2G, should be replaced
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/15505
Jenkins, retest 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 feature
enabled
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/15505
@kayousterhout It takes some time to update the test report.
---
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
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/15505
@kayousterhout
Test results have been updated:
| [SPARK-17931](https://github.com/witgo/spark/tree/SPARK-17931)
|https://github.com/apache/spark/commit
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/15505
Do not know which pr causes the run time of this test case to be reduced
from 21.764 s to 9.566 s.
---
If your project is set up for it, you can reply to this email and have your
reply appear
GitHub user witgo opened a pull request:
https://github.com/apache/spark/pull/17139
[WIP][SPARK-18890][CORE](try 3) Move task serialization from the
TaskSetManager to the CoarseGrainedSchedulerBackend
## What changes were proposed in this pull request?
See https
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103622493
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala ---
@@ -454,33 +452,15 @@ private[spark] class TaskSetManager
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103622591
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -621,6 +615,80 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103627459
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -621,6 +615,80 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103627873
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -621,6 +615,80 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103631254
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -621,6 +615,80 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103631300
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -621,6 +615,80 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103631294
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -621,6 +615,80 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103631260
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -621,6 +615,80 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103631249
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskDescription.scala ---
@@ -52,8 +55,26 @@ private[spark] class TaskDescription(
val
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103631278
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -621,6 +615,80 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103631373
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -195,6 +197,11 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103631322
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskDescription.scala ---
@@ -52,8 +55,26 @@ private[spark] class TaskDescription(
val
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103631360
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -621,6 +615,80 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103631354
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/local/LocalSchedulerBackend.scala
---
@@ -82,9 +88,15 @@ private[spark] class LocalEndpoint
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103631341
--- Diff: core/src/test/scala/org/apache/spark/executor/ExecutorSuite.scala
---
@@ -164,17 +164,18 @@ class ExecutorSuite extends SparkFunSuite
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103631404
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskDescription.scala ---
@@ -130,7 +152,7 @@ private[spark] object TaskDescription
GitHub user witgo opened a pull request:
https://github.com/apache/spark/pull/17116
[SPARK-18890][CORE](try 2) Move task serialization from the TaskSetManager
to the CoarseGrainedSchedulerBackend
## What changes were proposed in this pull request?
See https
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103636619
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -621,6 +615,80 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r103174824
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskDescription.scala ---
@@ -23,7 +23,10 @@ import java.util.Properties
import
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/14311
@rxin @ericl
This PR may cause the following code to throw an exception
```scala
private def getRemoteValues(blockId: BlockId): Option[BlockResult] = {
getRemoteBytes
GitHub user witgo opened a pull request:
https://github.com/apache/spark/pull/14977
[Test Only][not ready for review][SPARK-6235][CORE]Address various 2G limits
## What changes were proposed in this pull request?
### Design
Setup for eliminating the various 2G
GitHub user witgo opened a pull request:
https://github.com/apache/spark/pull/14995
[Test Only][not ready for review][SPARK-6235][CORE]Address various 2G
limits
## What changes were proposed in this pull request?
### motivation
The various 2G limit in Spark
Github user witgo closed the pull request at:
https://github.com/apache/spark/pull/14977
---
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
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/14995
retest 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 feature
enabled and wishes so
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/14995
Jenkins, retest 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 feature
enabled
Github user witgo closed the pull request at:
https://github.com/apache/spark/pull/14662
---
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
Github user witgo closed the pull request at:
https://github.com/apache/spark/pull/14751
---
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
Github user witgo closed the pull request at:
https://github.com/apache/spark/pull/14664
---
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
Github user witgo closed the pull request at:
https://github.com/apache/spark/pull/14647
---
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
Github user witgo closed the pull request at:
https://github.com/apache/spark/pull/14658
---
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
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/14995
Jenkins, retest 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 feature
enabled
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15408#discussion_r82537791
--- Diff:
core/src/main/java/org/apache/spark/io/NioBasedBufferedFileInputStream.java ---
@@ -0,0 +1,91 @@
+/*
+ * Licensed under the Apache License
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15408#discussion_r82537694
--- Diff:
core/src/main/java/org/apache/spark/io/NioBasedBufferedFileInputStream.java ---
@@ -0,0 +1,77 @@
+/*
+ * Licensed under the Apache License
GitHub user witgo opened a pull request:
https://github.com/apache/spark/pull/15830
[WIP]Upgrade netty to 4.0.42.Final
## What changes were proposed in this pull request?
One of the important changes for 4.0.42.Final is "Support any FileRegion
implementation when
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15830#discussion_r87327561
--- Diff: core/src/main/scala/org/apache/spark/util/Utils.scala ---
@@ -,6 +2223,9 @@ private[spark] object Utils extends Logging
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r8669
--- Diff: core/src/main/scala/org/apache/spark/executor/Executor.scala ---
@@ -486,7 +481,7 @@ private[spark] class Executor(
* Download any missing
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/15505
cc @rxin
---
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
GitHub user witgo opened a pull request:
https://github.com/apache/spark/pull/15512
The SerializerInstance instance used when deserializing a TaskResult is not
reused
## What changes were proposed in this pull request?
The following code is called when the DirectTaskResult
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15512#discussion_r83774753
--- Diff: core/src/main/scala/org/apache/spark/scheduler/TaskResult.scala
---
@@ -77,14 +78,12 @@ private[spark] class DirectTaskResult[T
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15512#discussion_r83774768
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskResultGetter.scala ---
@@ -84,6 +90,7 @@ private[spark] class TaskResultGetter(sparkEnv: SparkEnv
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/15512
I also think that the time saved is all the registration which can be
skipped, but did not verify.
---
If your project is set up for it, you can reply to this email and have your
reply appear
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/15505
@wzhfy
Ok, the code has been modified
---
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
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r89249175
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskDescription.scala ---
@@ -17,27 +17,178 @@
package org.apache.spark.scheduler
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r89249162
--- Diff:
core/src/test/scala/org/apache/spark/scheduler/TaskSetManagerSuite.scala ---
@@ -592,47 +579,6 @@ class TaskSetManagerSuite extends SparkFunSuite
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/14995
Thanks, I am glad to hear this, and I want to solve the issues of reading,
storing and transmitting data as much as possible.
---
If your project is set up for it, you can reply to this email
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r89483305
--- Diff:
core/src/test/scala/org/apache/spark/scheduler/TaskSchedulerImplSuite.scala ---
@@ -139,29 +139,6 @@ class TaskSchedulerImplSuite extends
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r89246343
--- Diff:
core/src/test/scala/org/apache/spark/scheduler/TaskSchedulerImplSuite.scala ---
@@ -139,29 +139,6 @@ class TaskSchedulerImplSuite extends
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r89246494
--- Diff:
mesos/src/test/scala/org/apache/spark/scheduler/cluster/mesos/MesosFineGrainedSchedulerBackendSuite.scala
---
@@ -345,7 +374,17 @@ class
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/14995
This PR is Test only, it used to
1. verify code through CI
2. verify the effectiveness of the solution
includes two underlying API changes.
1. Replace ByteBuffer
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/15830
@srowen thank you
---
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
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/14995
@srowen
This PR is a comprehensive solution. Used to solve address various 2G
limits, RPC memory footprint and other issues.
Users often encounter these problems. Why don't we need to solve
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/15505
ping @kayousterhout / @squito / @JoshRosen
---
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
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15297#discussion_r82922339
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/SkewShuffleRowRDD.scala
---
@@ -0,0 +1,147 @@
+/*
+ * Licensed to the Apache
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15297#discussion_r82922520
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/SkewShuffleRowRDD.scala
---
@@ -0,0 +1,147 @@
+/*
+ * Licensed to the Apache
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15297#discussion_r82742056
--- Diff: core/src/main/scala/org/apache/spark/MapOutputTracker.scala ---
@@ -138,13 +138,16 @@ private[spark] abstract class MapOutputTracker(conf:
SparkConf
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15297#discussion_r82743168
--- Diff: core/src/main/scala/org/apache/spark/shuffle/ShuffleManager.scala
---
@@ -48,7 +48,8 @@ private[spark] trait ShuffleManager {
handle
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15297#discussion_r82742902
--- Diff: core/src/main/scala/org/apache/spark/MapOutputTracker.scala ---
@@ -687,18 +691,21 @@ private[spark] object MapOutputTracker extends
Logging
GitHub user witgo opened a pull request:
https://github.com/apache/spark/pull/15505
[WIP][SPARK-17931]taskScheduler has some unneeded serialization
## What changes were proposed in this pull request?
When taskScheduler instantiates TaskDescription, it calls
Github user witgo commented on the issue:
https://github.com/apache/spark/pull/15505
@kayousterhout
Here are my thoughts:
move the serialization out of the `TaskSetManager.resourceOffer` method.
Split resourceOffer and serialization process, So that we can make
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r95723533
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskDescription.scala ---
@@ -52,7 +55,36 @@ private[spark] class TaskDescription(
val
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r95732717
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/cluster/CoarseGrainedSchedulerBackend.scala
---
@@ -245,6 +245,16 @@ class
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r95731120
--- Diff:
core/src/test/scala/org/apache/spark/scheduler/DAGSchedulerSuite.scala ---
@@ -517,6 +518,32 @@ class DAGSchedulerSuite extends SparkFunSuite
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r95731056
--- Diff:
core/src/test/scala/org/apache/spark/scheduler/DAGSchedulerSuite.scala ---
@@ -517,6 +518,32 @@ class DAGSchedulerSuite extends SparkFunSuite
Github user witgo commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r96104686
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskDescription.scala ---
@@ -52,7 +55,43 @@ private[spark] class TaskDescription(
val
701 - 800 of 866 matches
Mail list logo