Uploaded image for project: 'Hama'
  1. Hama
  2. HAMA-642

Make GraphRunner disk based

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Resolved
    • Major
    • Resolution: Won't Fix
    • 0.5.0
    • None
    • graph
    • None

    Description

      To improve scalability we can improve the graph runner to be disk based.

      Which basically means:

      • We have just a single Vertex instance that get's refilled.
      • We directly write vertices to disk after partitioning
      • In every superstep we iterate over the vertices on disk, fill the vertex instance and call the users compute functions

      Problems:

      • State other than vertex value can't be stored easy
      • How do we deal with random access after messages have arrived?

      So I think we should make the graph runner more hybrid, like using the queues we have implemented in the messaging. So the graphrunner can be configured to run completely on disk, in cached mode or in in-memory mode.

      Attachments

        1. HAMA-642_unix_1.patch
          760 kB
          Thomas Jungblut
        2. HAMA-642_unix_2.patch
          760 kB
          Thomas Jungblut
        3. HAMA-642_unix_3.patch
          762 kB
          Thomas Jungblut
        4. HAMA-642_unix_4.patch
          185 kB
          Thomas Jungblut
        5. HAMA-scale_1.patch
          501 kB
          Thomas Jungblut
        6. HAMA-scale_2.patch
          773 kB
          Thomas Jungblut
        7. HAMA-scale_3.patch
          780 kB
          Thomas Jungblut
        8. HAMA-scale_4.patch
          780 kB
          Thomas Jungblut

        Issue Links

          Activity

            People

              udanax Edward J. Yoon
              thomas.jungblut Thomas Jungblut
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: