Uploaded image for project: 'Apache Cassandra'
  1. Apache Cassandra
  2. CASSANDRA-13338

JMX: EstimatedPartitionCount is expensive

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Open
    • Normal
    • Resolution: Unresolved
    • None
    • Legacy/Observability
    • Low Hanging Fruit

    Description

      EstimatedPartitionCount / EstimatedRowCount / SnapshotSize seem particularly expensive. For example on our system org.apache.cassandra.metrics:type=ColumnFamily,name=SnapshotsSize can take as much as half a second. (This was resolved by CASSANDRA-18111).

      All this accumulated means that export stats for all your tables (with metrics-graphite or jmx_exporter) is going to take quite some time.

      We should certainly try to find the most expensive end points and see if there is a way to cache some of the values.

      cc: rgerard

      Attachments

        1. Throughput_vs_Threads_Combined _true_snapshot_size.png
          100 kB
          Stefan Miklosovic
        2. Throughput_vs_Threads_Combined_snapshot_listing.png
          102 kB
          Stefan Miklosovic
        3. Percentiles_vs_Threads_Combined_true_snapshot_size.png
          146 kB
          Stefan Miklosovic
        4. Percentiles_vs_Threads_Combined_snapshot_listing.png
          162 kB
          Stefan Miklosovic
        5. Average_Time_vs_Threads_Combined_true_snapshot_size.png
          101 kB
          Stefan Miklosovic
        6. Average_Time_vs_Threads_Combined_snapshot_listing.png
          112 kB
          Stefan Miklosovic

        Issue Links

          Activity

            People

              smiklosovic Stefan Miklosovic
              iksaif Corentin Chary
              Stefan Miklosovic
              Votes:
              0 Vote for this issue
              Watchers:
              6 Start watching this issue

              Dates

                Created:
                Updated: