Uploaded image for project: 'Hadoop YARN'
  1. Hadoop YARN
  2. YARN-6223 [Umbrella] Natively support GPU configuration/discovery/scheduling/isolation on YARN
  3. YARN-6620

Add support in NodeManager to isolate GPU devices by using CGroups

    XMLWordPrintableJSON

Details

    • Sub-task
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • None
    • 3.1.0
    • None
    • None
    • Reviewed

    Description

      This JIRA plan to add support of:
      1) GPU configuration for NodeManagers
      2) Isolation in CGroups. (Java side).
      3) NM restart and recovery allocated GPU devices

      Attachments

        1. YARN-6620.001.patch
          71 kB
          Wangda Tan
        2. YARN-6620.002.patch
          70 kB
          Wangda Tan
        3. YARN-6620.003.patch
          71 kB
          Wangda Tan
        4. YARN-6620.004.patch
          81 kB
          Wangda Tan
        5. YARN-6620.005.patch
          95 kB
          Wangda Tan
        6. YARN-6620.006-WIP.patch
          137 kB
          Wangda Tan
        7. YARN-6620.007.patch
          168 kB
          Wangda Tan
        8. YARN-6620.008.patch
          173 kB
          Wangda Tan
        9. YARN-6620.009.patch
          175 kB
          Wangda Tan
        10. YARN-6620.010.patch
          176 kB
          Wangda Tan
        11. YARN-6620.011.patch
          177 kB
          Wangda Tan
        12. YARN-6620.012.patch
          178 kB
          Wangda Tan
        13. YARN-6620.013.patch
          178 kB
          Wangda Tan
        14. YARN-6620.014.patch
          182 kB
          Wangda Tan
        15. YARN-6620.015.patch
          184 kB
          Wangda Tan
        16. YARN-6620.016.patch
          184 kB
          Wangda Tan
        17. YARN-6620.017.patch
          184 kB
          Wangda Tan

        Issue Links

          breaks

          Bug - A problem which impairs or prevents the functions of the product. YARN-7396 NPE when accessing container logs due to null dirsHandler

          • Major - Major loss of function.
          • Resolved
          depends upon

          Sub-task - The sub-task of the issue YARN-6852 [YARN-6223] Native code changes to support isolate GPU devices by using CGroups

          • Major - Major loss of function.
          • Resolved

          Improvement - An improvement or enhancement to an existing feature or task. YARN-7033 Add support for NM Recovery of assigned resources (e.g. GPU's, NUMA, FPGA's) to container

          • Major - Major loss of function.
          • Resolved
          relates to

          Sub-task - The sub-task of the issue YARN-9280 Backport YARN-6620 to YARN-8200/branch-2

          • Major - Major loss of function.
          • Resolved

          Activity

            People

              leftnoteasy Wangda Tan
              leftnoteasy Wangda Tan
              Votes:
              0 Vote for this issue
              Watchers:
              19 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: