Uploaded image for project: 'Hadoop Common'
  1. Hadoop Common
  2. HADOOP-2758

Reduce memory copies when data is read from DFS

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Closed
    • Major
    • Resolution: Fixed
    • None
    • 0.17.0
    • None
    • None
    • DataNode takes 50% less CPU while serving data to clients.

    Description

      Currently datanode and client part of DFS perform multiple copies of data on the 'read path' (i.e. path from storage on datanode to user buffer on the client). This jira reduces these copies by enhancing data read protocol and implementation of read on both datanode and the client. I will describe the changes in next comment.

      Requirement is that this fix should reduce CPU used and should not cause regression in any benchmarks. It might not improve the benchmarks since most benchmarks are not cpu bound.

      Attachments

        1. HADOOP-2758.patch
          33 kB
          Raghu Angadi
        2. HADOOP-2758.patch
          33 kB
          Raghu Angadi
        3. HADOOP-2758.patch
          33 kB
          Raghu Angadi
        4. HADOOP-2758.patch
          33 kB
          Raghu Angadi
        5. HADOOP-2758.patch
          28 kB
          Raghu Angadi
        6. HADOOP-2758.patch
          21 kB
          Raghu Angadi

        Issue Links

          Activity

            People

              rangadi Raghu Angadi
              rangadi Raghu Angadi
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: