Thank you!

I know it is time consuming to run all these benchmarks and objectively publish the results. I desperately wanted to know how the JDK 5.0 compared to other VMs as well, but I simply didn't have the time to run all these benchmarks myself. Thanks again!

[EMAIL PROTECTED] wrote:
I have updated a web page showing results of performance measurement of
various Java runtimes.

  Performance Comparison of Java/.NET Runtimes (Oct 2004)
  http://www.shudo.net/jit/perf/

The benchmarks on the page are mainly compute intensive and not
server-side ones: SPEC JVM98, SciMark 2.0, Linpack benchmark and
Eratosthenes sieve.  Such benchmark results seem not to be in much
demand these days but recently released JDK 5.0 pushed me into doing
performance measurement.

This time Opteron processor (with 64 bit and 32 bit binaries) was
tested in addition to Pentium 4.  Another interesting point is .NET
Framework, which was tested with simple benchmarks ported to C#.

Possibly interesting points are:

  - C#/.NET performance

    - Startup process and/or JIT compilation may be heavier than
      Java runtimes including HotSpot Server VM.

    - Throughput of .NET Framework is pretty high.
      C#/.NET shown higher performance than HotSpot Server VM
      in Linpack benchmark with large data.

  - Difference between 64 bit (AMD64) code and 32 bit (x86) code on Opteron

    - 64 bit code is faster in some benchmarks, and in other
      benchmarks 32 bit code is faster.

    - 64 bit code should pay the penalty of larger pointer (reference)
      and benefit (a little ?) from faster 64 bit integer arithmetic.

Hope you enjoy,

  Kazuyuki Shudo        [EMAIL PROTECTED]       http://www.shudo.net/


---------------------------------------------------------------------- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]




-- Matthew Avery Senior Developer (513) 470-5316 http://www.einnovation.com/


---------------------------------------------------------------------- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Reply via email to