Hadoop HDFS
  1. Hadoop HDFS
  2. HDFS-6599

2.4 addBlock is 10 to 20 times slower compared to 0.23

    Details

    • Type: Bug Bug
    • Status: Closed
    • Priority: Blocker Blocker
    • Resolution: Fixed
    • Affects Version/s: 2.3.0, 2.4.0
    • Fix Version/s: 2.5.0
    • Component/s: None
    • Labels:
      None
    • Target Version/s:
    • Hadoop Flags:
      Reviewed

      Description

      From one of our busiest 0.23 clusters:

      "AddBlockAvgTime" : 0.9514711501719515
      "CreateAvgTime" : 1.7564162389174
      "CompleteAvgTime" : 1.3310406035056548
      "BlockReceivedAndDeletedAvgTime" : 0.661210005151392

      From a not-so-busy 2.4 cluster:

      "AddBlockAvgTime" : 10.083333333333334
      "CreateAvgTime" : 1.0
      "CompleteAvgTime" : 1.1111111111111112
      "BlockReceivedAndDeletedAvgTime" : 0.07692307692307694

      When the 2.4 cluster gets a moderate amount of write requests, the latency is terrible. E.g. addBlock goes upward of 60ms.

      1. HDFS-6599.patch
        20 kB
        Daryn Sharp

        Issue Links

          Activity

          Hide
          Kihwal Lee added a comment -

          This is caused by a change in BlockPlacementPolicyDefault#isGoodTarget(). The synchronization overhead in HeartBeatManager or DatanodeManager might also be worse.

          In 0.23 it didn't go there at all. It just looked at the topology.
          int size = clusterMap.getNumOfLeaves()

          In 2.4.x, it is calling this:
          int size = stats.getNumDatanodesInService()

          Since the network topology doesn't tell you the accurate state of the cluster, it is a right thing to do semantically. But its performance impact should have been considered. This change was made in HDFS-5837.

          Show
          Kihwal Lee added a comment - This is caused by a change in BlockPlacementPolicyDefault#isGoodTarget() . The synchronization overhead in HeartBeatManager or DatanodeManager might also be worse. In 0.23 it didn't go there at all. It just looked at the topology. int size = clusterMap.getNumOfLeaves() In 2.4.x, it is calling this: int size = stats.getNumDatanodesInService() Since the network topology doesn't tell you the accurate state of the cluster, it is a right thing to do semantically. But its performance impact should have been considered. This change was made in HDFS-5837 .
          Hide
          Daryn Sharp added a comment -

          The problem is that calls to DataNodeManager#getDatanodeListForReport are not cheap. Prior to the linked bug, nothing but metrics invoked the expensive routines. I've tweaked the HeartbeatManager to track the in-service nodes and xceivers so the load check is cheap.

          Show
          Daryn Sharp added a comment - The problem is that calls to DataNodeManager#getDatanodeListForReport are not cheap. Prior to the linked bug, nothing but metrics invoked the expensive routines. I've tweaked the HeartbeatManager to track the in-service nodes and xceivers so the load check is cheap.
          Hide
          Hadoop QA added a comment -

          +1 overall. Here are the results of testing the latest attachment
          http://issues.apache.org/jira/secure/attachment/12654372/HDFS-6599.patch
          against trunk revision .

          +1 @author. The patch does not contain any @author tags.

          +1 tests included. The patch appears to include 2 new or modified test files.

          +1 javac. The applied patch does not increase the total number of javac compiler warnings.

          +1 javadoc. There were no new javadoc warning messages.

          +1 eclipse:eclipse. The patch built with eclipse:eclipse.

          +1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.

          +1 release audit. The applied patch does not increase the total number of release audit warnings.

          +1 core tests. The patch passed unit tests in hadoop-hdfs-project/hadoop-hdfs.

          +1 contrib tests. The patch passed contrib unit tests.

          Test results: https://builds.apache.org/job/PreCommit-HDFS-Build/7286//testReport/
          Console output: https://builds.apache.org/job/PreCommit-HDFS-Build/7286//console

          This message is automatically generated.

          Show
          Hadoop QA added a comment - +1 overall . Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12654372/HDFS-6599.patch against trunk revision . +1 @author . The patch does not contain any @author tags. +1 tests included . The patch appears to include 2 new or modified test files. +1 javac . The applied patch does not increase the total number of javac compiler warnings. +1 javadoc . There were no new javadoc warning messages. +1 eclipse:eclipse . The patch built with eclipse:eclipse. +1 findbugs . The patch does not introduce any new Findbugs (version 1.3.9) warnings. +1 release audit . The applied patch does not increase the total number of release audit warnings. +1 core tests . The patch passed unit tests in hadoop-hdfs-project/hadoop-hdfs. +1 contrib tests . The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-HDFS-Build/7286//testReport/ Console output: https://builds.apache.org/job/PreCommit-HDFS-Build/7286//console This message is automatically generated.
          Hide
          Aaron T. Myers added a comment -

          +1, the latest patch looks good to me.

          Thanks for taking care of this, Daryn.

          Show
          Aaron T. Myers added a comment - +1, the latest patch looks good to me. Thanks for taking care of this, Daryn.
          Hide
          Vinayakumar B added a comment -

          Changes looks good.

          How to get those statistics? to test with patch applied.

          Show
          Vinayakumar B added a comment - Changes looks good. How to get those statistics? to test with patch applied.
          Hide
          Daryn Sharp added a comment -

          Thanks Aaron, I'll check in shortly. Vinayakumar, the problem doesn't become evident until you have a multi-thousand node cluster.

          Show
          Daryn Sharp added a comment - Thanks Aaron, I'll check in shortly. Vinayakumar, the problem doesn't become evident until you have a multi-thousand node cluster.
          Hide
          Hudson added a comment -

          FAILURE: Integrated in Hadoop-trunk-Commit #5915 (See https://builds.apache.org/job/Hadoop-trunk-Commit/5915/)
          HDFS-6599. 2.4 addBlock is 10 to 20 times slower compared to 0.23 (daryn) (daryn: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1611737)

          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/protocol/DatanodeInfo.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/BlockPlacementPolicyDefault.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeManager.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeStatistics.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/HeartbeatManager.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSClusterStats.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSNamesystem.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/blockmanagement/TestReplicationPolicyConsiderLoad.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/TestNamenodeCapacityReport.java
          Show
          Hudson added a comment - FAILURE: Integrated in Hadoop-trunk-Commit #5915 (See https://builds.apache.org/job/Hadoop-trunk-Commit/5915/ ) HDFS-6599 . 2.4 addBlock is 10 to 20 times slower compared to 0.23 (daryn) (daryn: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1611737 ) /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/protocol/DatanodeInfo.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/BlockPlacementPolicyDefault.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeManager.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeStatistics.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/HeartbeatManager.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSClusterStats.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSNamesystem.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/blockmanagement/TestReplicationPolicyConsiderLoad.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/TestNamenodeCapacityReport.java
          Hide
          Hudson added a comment -

          FAILURE: Integrated in Hadoop-Yarn-trunk #617 (See https://builds.apache.org/job/Hadoop-Yarn-trunk/617/)
          HDFS-6599. 2.4 addBlock is 10 to 20 times slower compared to 0.23 (daryn) (daryn: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1611737)

          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/protocol/DatanodeInfo.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/BlockPlacementPolicyDefault.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeManager.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeStatistics.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/HeartbeatManager.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSClusterStats.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSNamesystem.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/blockmanagement/TestReplicationPolicyConsiderLoad.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/TestNamenodeCapacityReport.java
          Show
          Hudson added a comment - FAILURE: Integrated in Hadoop-Yarn-trunk #617 (See https://builds.apache.org/job/Hadoop-Yarn-trunk/617/ ) HDFS-6599 . 2.4 addBlock is 10 to 20 times slower compared to 0.23 (daryn) (daryn: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1611737 ) /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/protocol/DatanodeInfo.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/BlockPlacementPolicyDefault.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeManager.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeStatistics.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/HeartbeatManager.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSClusterStats.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSNamesystem.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/blockmanagement/TestReplicationPolicyConsiderLoad.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/TestNamenodeCapacityReport.java
          Hide
          Hudson added a comment -

          FAILURE: Integrated in Hadoop-Hdfs-trunk #1809 (See https://builds.apache.org/job/Hadoop-Hdfs-trunk/1809/)
          HDFS-6599. 2.4 addBlock is 10 to 20 times slower compared to 0.23 (daryn) (daryn: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1611737)

          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/protocol/DatanodeInfo.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/BlockPlacementPolicyDefault.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeManager.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeStatistics.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/HeartbeatManager.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSClusterStats.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSNamesystem.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/blockmanagement/TestReplicationPolicyConsiderLoad.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/TestNamenodeCapacityReport.java
          Show
          Hudson added a comment - FAILURE: Integrated in Hadoop-Hdfs-trunk #1809 (See https://builds.apache.org/job/Hadoop-Hdfs-trunk/1809/ ) HDFS-6599 . 2.4 addBlock is 10 to 20 times slower compared to 0.23 (daryn) (daryn: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1611737 ) /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/protocol/DatanodeInfo.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/BlockPlacementPolicyDefault.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeManager.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeStatistics.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/HeartbeatManager.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSClusterStats.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSNamesystem.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/blockmanagement/TestReplicationPolicyConsiderLoad.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/TestNamenodeCapacityReport.java
          Hide
          Hudson added a comment -

          SUCCESS: Integrated in Hadoop-Mapreduce-trunk #1836 (See https://builds.apache.org/job/Hadoop-Mapreduce-trunk/1836/)
          HDFS-6599. 2.4 addBlock is 10 to 20 times slower compared to 0.23 (daryn) (daryn: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1611737)

          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/protocol/DatanodeInfo.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/BlockPlacementPolicyDefault.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeManager.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeStatistics.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/HeartbeatManager.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSClusterStats.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSNamesystem.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/blockmanagement/TestReplicationPolicyConsiderLoad.java
          • /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/TestNamenodeCapacityReport.java
          Show
          Hudson added a comment - SUCCESS: Integrated in Hadoop-Mapreduce-trunk #1836 (See https://builds.apache.org/job/Hadoop-Mapreduce-trunk/1836/ ) HDFS-6599 . 2.4 addBlock is 10 to 20 times slower compared to 0.23 (daryn) (daryn: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1611737 ) /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/protocol/DatanodeInfo.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/BlockPlacementPolicyDefault.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeManager.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/DatanodeStatistics.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/HeartbeatManager.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSClusterStats.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSNamesystem.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/blockmanagement/TestReplicationPolicyConsiderLoad.java /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/namenode/TestNamenodeCapacityReport.java
          Hide
          Vinayakumar B added a comment -

          Vinayakumar, the problem doesn't become evident until you have a multi-thousand node cluster.

          Thanks Daryn Sharp, I am just curious to know, how you collected those statistics.

          Show
          Vinayakumar B added a comment - Vinayakumar, the problem doesn't become evident until you have a multi-thousand node cluster. Thanks Daryn Sharp , I am just curious to know, how you collected those statistics.
          Hide
          Anthony Hsu added a comment -

          We're encountering NameNode slowness on Hadoop 2.3 and wondering whether this patch will solve the problem. Is there any way to benchmark on a smaller cluster? Why does this problem only become evident on large, multi-thousand node clusters?

          Show
          Anthony Hsu added a comment - We're encountering NameNode slowness on Hadoop 2.3 and wondering whether this patch will solve the problem. Is there any way to benchmark on a smaller cluster? Why does this problem only become evident on large, multi-thousand node clusters?
          Hide
          Kihwal Lee added a comment -

          There have been many performance improvements since 2.3. Examples: HDFS-7097 and HDFS-7615 fix writers being unfairly penalized. HDFS-7217 reduces the write locking of namesystem by 30-40%. This is a big deal for busy clusters. If you use the audit logging, try async audit logging (HDFS-5241). This should be already in 2.3.

          Show
          Kihwal Lee added a comment - There have been many performance improvements since 2.3. Examples: HDFS-7097 and HDFS-7615 fix writers being unfairly penalized. HDFS-7217 reduces the write locking of namesystem by 30-40%. This is a big deal for busy clusters. If you use the audit logging, try async audit logging ( HDFS-5241 ). This should be already in 2.3.
          Hide
          Anthony Hsu added a comment -

          Thanks for the pointers, Kihwal. We'll take a look at some of those patches and consider applying them.

          BTW, we applied the patch for this ticket and noticed a significant improvement in NameNode latency and throughput!

          Show
          Anthony Hsu added a comment - Thanks for the pointers, Kihwal. We'll take a look at some of those patches and consider applying them. BTW, we applied the patch for this ticket and noticed a significant improvement in NameNode latency and throughput!

            People

            • Assignee:
              Daryn Sharp
              Reporter:
              Kihwal Lee
            • Votes:
              0 Vote for this issue
              Watchers:
              22 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved:

                Development