Details

    • Type: Task
    • Status: Closed (View Workflow)
    • Priority: Major
    • Resolution: Done
    • Affects Version/s: 1.4.0 beta
    • Fix Version/s: 1.4.0 final
    • Component/s: general
    • Labels:
      None
    • Environment:

      Tested on JBR 1.2.0, Java 1.5.0_03, Windows XP, P4 2,6GHz, 1GB RAM

      Description

      The JBR framework adds a bit of overhead to the overall transport performance. A simple round-trip invocation test executed 1 000, 10 000 and 100 000 times sequentially using RMI on localhost compared to a simple stand-alone RMI server and client generated the following results. Even though the test was performed using RMI it is most likely the same for all supported protocols.

      RMI using JBR (stand-alone server based on the SimpleServer example)
      1 000 ~1,3 sec
      10 000 ~6,4 sec
      100 000 ~55 sec

      RMI Stand-alone
      1 000 ~0.5 sec
      10 000 ~2.5 sec
      100 000 ~20 sec

      JBR seems to add a 2.5 times overhead.

        Gliffy Diagrams

          Attachments

            Activity

              People

              • Assignee:
                tom.elrod Tom Elrod
                Reporter:
                mafor Mats Forslöf
              • Votes:
                0 Vote for this issue
                Watchers:
                0 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved: