[GitHub] carbondata pull request #2953: [CARBONDATA-3132]Correct the task disrtibutio...

[qiuchenjian-2]

GitHub user akashrn5 opened a pull request:

    https://github.com/apache/carbondata/pull/2953

    [CARBONDATA-3132]Correct the task disrtibution in case of compaction when the actual block nodes and active nodes are different

    ### Why This PR?
    There is an unequal distribution of tasks during compaction
    ex: When the load is done using replication factor 2 and all nodes are active and during compaction one node is down, basically it is not active executor, so the task distribution should take care to distribute the tasks equally among all the active executors instead of giving more tasks to single executor and less to other executor. But sometimes the unequal distribution happens and the compaction becomes sow.
   
    ### Solution
    Currently we are not getting active executors before the node block mapping and sending the list of active executors as null, which will lead to the above problem sometimes. so get the active executors and send for node block mapping logic which will handle to distribute equally.
   
   
   
    Be sure to do all of the following checklist to help us incorporate
    your contribution quickly and easily:
   
     - [x] Any interfaces changed?
     NA
     - [x] Any backward compatibility impacted?
     NA
     - [x] Document update required?
    NA
   
     - [x] Testing done
    tested using 3 node and 6 node cluster
            Please provide details on
            - Whether new unit test cases have been added or why no new tests are required?
            - How it is tested? Please attach test report.
            - Is it a performance related change? Please attach the performance test report.
            - Any additional information to help reviewers in testing this change.
           
     - [x] For large changes, please consider breaking it into sub-tasks under an umbrella JIRA.
    NA


You can merge this pull request into a Git repository by running:

    $ git pull https://github.com/akashrn5/incubator-carbondata distribute

Alternatively you can review and apply these changes as the patch at:

    https://github.com/apache/carbondata/pull/2953.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

    This closes #2953
   
----
commit caaa044b461a64f28be6107e55e8ab2a1ae2bf4d
Author: brijoobopanna <brijoobopanna@...>
Date:   2018-11-27T06:38:15Z

    Correct the task disrtibution in case of compaction when the actual block nodes and active nodes are different

----


---

[qiuchenjian-2]

Github user akashrn5 commented on a diff in the pull request:

    https://github.com/apache/carbondata/pull/2953#discussion_r236537747
 
    --- Diff: tools/cli/src/test/java/org/apache/carbondata/tool/CarbonCliTest.java ---
    @@ -205,7 +206,7 @@ public void testSummaryOutputAll() {
         expectedOutput = buildLines(
             "## version Details",
             "written_by  Version         ",
    -        "TestUtil    1.6.0-SNAPSHOT  ");
    +        "TestUtil    "+ CarbonVersionConstants.CARBONDATA_VERSION+"  ");
    --- End diff --
   
    removed the harded value here


---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2953
 
    Build Success with Spark 2.1.0, Please check CI http://136.243.101.176:8080/job/ApacheCarbonPRBuilder2.1/1544/



---

[qiuchenjian-2]

Github user qiuchenjian commented on a diff in the pull request:

    https://github.com/apache/carbondata/pull/2953#discussion_r236543235
 
    --- Diff: processing/src/main/java/org/apache/carbondata/processing/util/CarbonLoaderUtil.java ---
    @@ -536,7 +537,8 @@ public static Dictionary getDictionary(AbsoluteTableIdentifier absoluteTableIden
        */
       public static Map<String, List<Distributable>> nodeBlockMapping(List<Distributable> blockInfos) {
         // -1 if number of nodes has to be decided based on block location information
    -    return nodeBlockMapping(blockInfos, -1);
    +    return nodeBlockMapping(blockInfos, -1, null,
    --- End diff --
   
    the choice of BlockAssignmentStrategy should be judged by  CarbonProperties.getInstance().isLoadSkewedDataOptimizationEnabled() ?


---

[qiuchenjian-2]

Github user akashrn5 commented on a diff in the pull request:

    https://github.com/apache/carbondata/pull/2953#discussion_r236551537
 
    --- Diff: processing/src/main/java/org/apache/carbondata/processing/util/CarbonLoaderUtil.java ---
    @@ -536,7 +537,8 @@ public static Dictionary getDictionary(AbsoluteTableIdentifier absoluteTableIden
        */
       public static Map<String, List<Distributable>> nodeBlockMapping(List<Distributable> blockInfos) {
         // -1 if number of nodes has to be decided based on block location information
    -    return nodeBlockMapping(blockInfos, -1);
    +    return nodeBlockMapping(blockInfos, -1, null,
    --- End diff --
   
    yes, in compaction case `BlockAssignmentStrategy.BLOCK_NUM_FIRST` is default , same as before


---

[qiuchenjian-2]

Github user akashrn5 commented on a diff in the pull request:

    https://github.com/apache/carbondata/pull/2953#discussion_r236552294
 
    --- Diff: processing/src/main/java/org/apache/carbondata/processing/util/CarbonLoaderUtil.java ---
    @@ -536,7 +537,8 @@ public static Dictionary getDictionary(AbsoluteTableIdentifier absoluteTableIden
        */
       public static Map<String, List<Distributable>> nodeBlockMapping(List<Distributable> blockInfos) {
         // -1 if number of nodes has to be decided based on block location information
    -    return nodeBlockMapping(blockInfos, -1);
    +    return nodeBlockMapping(blockInfos, -1, null,
    --- End diff --
   
    here i just did the refactoring


---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2953
 
    Build Success with Spark 2.2.1, Please check CI http://95.216.28.178:8080/job/ApacheCarbonPRBuilder1/1756/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2953
 
    Build Success with Spark 2.1.0, Please check CI http://136.243.101.176:8080/job/ApacheCarbonPRBuilder2.1/1546/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2953
 
    Build Success with Spark 2.3.1, Please check CI http://136.243.101.176:8080/job/carbondataprbuilder2.3/9806/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2953
 
    Build Success with Spark 2.2.1, Please check CI http://95.216.28.178:8080/job/ApacheCarbonPRBuilder1/1758/



---

[qiuchenjian-2]

Github user ravipesala commented on a diff in the pull request:

    https://github.com/apache/carbondata/pull/2953#discussion_r236640610
 
    --- Diff: integration/spark-common/src/main/scala/org/apache/carbondata/spark/rdd/CarbonMergerRDD.scala ---
    @@ -401,7 +401,11 @@ class CarbonMergerRDD[K, V](
               .add(new CarbonInputSplitTaskInfo(entry._1, entry._2).asInstanceOf[Distributable])
         )
     
    -    val nodeBlockMap = CarbonLoaderUtil.nodeBlockMapping(taskInfoList, -1)
    +    // get all the active nodes of cluster and prepare the nodeBlockMap based on these nodes
    +    val activeNodes = DistributionUtil
    +      .ensureExecutorsAndGetNodeList(taskInfoList.asScala, sparkContext)
    +
    +    val nodeBlockMap = CarbonLoaderUtil.nodeBlockMapping(taskInfoList, -1, activeNodes.asJava)
    --- End diff --
   
    Below code is redundant, please remove it


---

[qiuchenjian-2]

Github user akashrn5 commented on a diff in the pull request:

    https://github.com/apache/carbondata/pull/2953#discussion_r236646439
 
    --- Diff: integration/spark-common/src/main/scala/org/apache/carbondata/spark/rdd/CarbonMergerRDD.scala ---
    @@ -401,7 +401,11 @@ class CarbonMergerRDD[K, V](
               .add(new CarbonInputSplitTaskInfo(entry._1, entry._2).asInstanceOf[Distributable])
         )
     
    -    val nodeBlockMap = CarbonLoaderUtil.nodeBlockMapping(taskInfoList, -1)
    +    // get all the active nodes of cluster and prepare the nodeBlockMap based on these nodes
    +    val activeNodes = DistributionUtil
    +      .ensureExecutorsAndGetNodeList(taskInfoList.asScala, sparkContext)
    +
    +    val nodeBlockMap = CarbonLoaderUtil.nodeBlockMapping(taskInfoList, -1, activeNodes.asJava)
    --- End diff --
   
    done


---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2953
 
    Build Failed  with Spark 2.1.0, Please check CI http://136.243.101.176:8080/job/ApacheCarbonPRBuilder2.1/1554/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2953
 
    Build Failed  with Spark 2.3.1, Please check CI http://136.243.101.176:8080/job/carbondataprbuilder2.3/9813/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2953
 
    Build Failed with Spark 2.2.1, Please check CI http://95.216.28.178:8080/job/ApacheCarbonPRBuilder1/1765/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2953
 
    Build Success with Spark 2.1.0, Please check CI http://136.243.101.176:8080/job/ApacheCarbonPRBuilder2.1/1555/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2953
 
    Build Success with Spark 2.3.1, Please check CI http://136.243.101.176:8080/job/carbondataprbuilder2.3/9814/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2953
 
    Build Success with Spark 2.2.1, Please check CI http://95.216.28.178:8080/job/ApacheCarbonPRBuilder1/1766/



---

[qiuchenjian-2]

Github user ravipesala commented on the issue:

    https://github.com/apache/carbondata/pull/2953
 
    LGTM


---

[qiuchenjian-2]

Github user asfgit closed the pull request at:

    https://github.com/apache/carbondata/pull/2953


---