Uploaded image for project: 'WiredTiger'
  1. WiredTiger
  2. WT-7041

Get incremental distribution for _id for large scale workload

    • Type: Icon: Improvement Improvement
    • Resolution: Fixed
    • Priority: Icon: Major - P3 Major - P3
    • WT10.0.0
    • Affects Version/s: None
    • Component/s: None
    • Labels:
      None
    • 8
    • Storage - Ra 2021-01-11

      While working on WT-6967 I figured that the new workload ScanWithLongLived.yml has a load phase that creates documents with the default _id. The default _id is of ObjectID type. The read and update phases use a filter on the _id to distribute the read/write load binomially across the key range. The filter gets created assuming that the _id kets are of numeric type. The test has assumed that _id would be numerically incrementing in a collection, which is actually not true. This leads to read/update actors matching no documents at all.

      The task of this ticket is to fix the workload. An initial starting point could be to get the _id populated in a numerically incremental way. An alternate needs to be found if that ends up being challenging or we might even need to introduce such a capability in genny itself.

            Assignee:
            alex.cameron@mongodb.com Alex Cameron (Inactive)
            Reporter:
            sulabh.mahajan@mongodb.com Sulabh Mahajan
            Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

              Created:
              Updated:
              Resolved: