Uploaded image for project: 'Geode'
  1. Geode
  2. GEODE-3701

Improve docs on hash index performance

Attach filesAttach ScreenshotVotersWatch issueWatchersCreate sub-taskLinkCloneUpdate Comment AuthorReplace String in CommentUpdate Comment VisibilityDelete Comments
    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Closed
    • Major
    • Resolution: Fixed
    • None
    • 1.3.0, 1.4.0
    • None
    • None

    Description

      Stronger language is warranted for hash indexes.
      In a simple test, there was an order of magnitude difference in load and GII performance between the two types of indexes.

      Range:
      Client: Loaded 2,000,000 entries in 20,487 ms
      Server 1 Size: 1,272,322,680 bytes
      Server 2: GIIed in 12,523 ms
      Server 2 Size: 1,272,088,544 bytes

      Hash:
      Client: Loaded 2,000,000 entries in 257,016 ms
      Server 1 Size: 1,228,010,280 bytes
      Server 2: GIIed in 218,983 ms
      Server 2 Size: 1,228,161,664 bytes
      So, the difference in size was ~45MB, but the difference in time was an order of magnitude.

      Better wording suggested for file developing/query_index/creating_hash_indexes.html:
      Using hash index will degrade put performance and recovery time substantially. If memory is not a concern, it is recommended that you use range index.

      Attachments

        Activity

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

          People

            karensmolermiller Karen Smoler Miller
            karensmolermiller Karen Smoler Miller
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved:

              Issue deployment