Uploaded image for project: 'HBase'
  1. HBase
  2. HBASE-2066

Perf: parallelize puts

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Closed
    • Major
    • Resolution: Fixed
    • 0.20.2, 0.20.3
    • 0.90.0
    • None
    • None

    Description

      Right now with large region count tables, the write buffer is not efficient. This is because we issue potentially N RPCs, where N is the # of regions in the table. When N gets large (lets say 1200+) things become sloowwwww.

      Instead if we batch things up using a different RPC and use thread pools, we could see higher performance!

      This requires a RPC change...

      Attachments

        1. HBASE-2066-20-branch.txt
          32 kB
          ryan rawson
        2. HBASE-2066-3.patch
          31 kB
          ryan rawson
        3. HBASE-2066-branch.patch
          22 kB
          ryan rawson
        4. HBASE-2066-v2.patch
          25 kB
          ryan rawson
        5. TestBatchPut.java
          2 kB
          Cosmin Lehene

        Issue Links

          is related to

          New Feature - A new feature of the product, which has yet to be developed. HBASE-1845 MultiGet, MultiDelete, and MultiPut - batched to the appropriate region servers

          • Major - Major loss of function.
          • Closed

          Activity

            People

              ryanobjc ryan rawson
              ryanobjc ryan rawson
              Votes:
              0 Vote for this issue
              Watchers:
              5 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: