Uploaded image for project: 'Hadoop HDFS'
  1. Hadoop HDFS
  2. HDFS-13694

Making md5 computing being in parallel with image loading

Log workAgile BoardRank to TopRank to BottomAttach filesAttach ScreenshotBulk Copy AttachmentsBulk Move AttachmentsVotersWatch issueWatchersCreate sub-taskConvert to sub-taskMoveLinkCloneLabelsUpdate Comment AuthorReplace String in CommentUpdate Comment VisibilityDelete Comments
    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • None
    • 2.10.0, 3.0.4, 3.3.0, 3.2.1, 2.9.3, 3.1.3
    • None
    • None
    • Reviewed

    Description

      During namenode image loading, it firstly compute the md5 and then load the image. Actually these two steps can be in parallel.

       Test this patch against a fsimage of a 70PB 2.4 cluster (200million files and 300million blocks), the image loading time be reduced from 1210 seconds to 1105 seconds.So it can reduce up to about 10% of time.

       

      Attachments

        1. HDFS-13694-001.patch
          2 kB
          zhouyingchao
        2. HDFS-13694-002.patch
          2 kB
          Lisheng Sun
        3. HDFS-13694-003.patch
          2 kB
          Lisheng Sun
        4. HDFS-13694-004.patch
          2 kB
          Lisheng Sun
        5. HDFS-13694-005.patch
          3 kB
          Lisheng Sun
        6. HDFS-13694-006.patch
          3 kB
          Lisheng Sun
        7. HDFS-13694-007.patch
          3 kB
          Íñigo Goiri

        Issue Links

        Activity

          This comment will be Viewable by All Users Viewable by All Users
          Cancel

          People

            leosun08 Lisheng Sun Assign to me
            sinago zhouyingchao
            Votes:
            0 Vote for this issue
            Watchers:
            15 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved:

              Slack

                Issue deployment