Add throughput probing state metrics

XMLWordPrintableJSON

    • Type: Improvement
    • Resolution: Fixed
    • Priority: Major - P3
    • 8.1.0-rc0
    • Affects Version/s: None
    • Component/s: None
    • None
    • Workload Scheduling
    • Fully Compatible
    • Workload Scheduling 2025-02-03, Workload Scheduling 2025-02-17
    • None
    • 3
    • None
    • None
    • None
    • None
    • None
    • None
    • None

      The throughput probing implementation currently reports the following metrics:

      • timesIncreased
      • totalAmountIncreased
      • timesDecreased
      • totalAmountDecreased
      • resizeDurationMicros

      Adding a few more metrics would help us better diagnose the behavior of throughput probing in tickets like HELP-69090:

      • timesProbedStable: the cumulative number of times the algorithm performed a stable probe
      • timesProbedUp: the cumulative number of times the algorithm probed up
      • timesProbedDown: the cumulative number of times the algorithm probed down

      Note that these metrics are different from timesIncreased and timesDecreased. The existing metrics measure how many times a probe up/down resulted in accepting the new concurrency.

              Assignee:
              Dominic Hernandez (Inactive)
              Reporter:
              Vojislav Stojkovic
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

                Created:
                Updated:
                Resolved: