Popular @
Forum Controls
Spotlight Features
The Rich Engineering Heritage Behind Dependency Injection
Andrew McVeigh takes us on a tour of the rich heritage behind dependency injection, what it represents, and tells us why its here to stay.
NetBeans 6: Matisse Updates
NetBeans 6 delivers great updates to the Matisse GUI builder. Spend a few minutes with Roman Strobl and get an expert briefing on what's new and what has changed. (sponsored)
Introduction to Groovy Part 3
In this, the third and final installation of Andres' Introduction to Groovy series, you learn about how Groovy handles variable numbers of arguments, named parameters, currying, and more about Groovy operators. Including, some new operators.
Easier Custom Components with Swing Fuse
Swing Fuse (actually just Fuse), is a framework designed to make it easier to create your own custom desktop components. In this article, Daniel Spiewak shows you how to get started and provides sample source code you can download.
Benchmark Analysis: Guice vs Spring
Willam Louth shows how he uses JXInsight Probes to investigate probable performance issues with code bases that he is not familiar with. He also highlights possible pitfalls in creating a benchmark, as well as in the analysis of results.
Replies:
9 -
Pages:
1
BEA JRockit 5.0 Sets New Performance Record
URL: LinuxElectrons
At 11:56 AM on Dec 12, 2005, Michael Urban wrote:
Fresh Jobs for Developers Post a job opportunity
The result was obtained using a Fujitsu PRIMEQUEST 480 server with 32 Intel® Itanium® 2 processors and the JRockit JVM 5.0. The result was the best 32-way score on non-SMT/HT hardware. It was also better than any result produced by Sun's HotSpot JVM on a machine of the same configuration.
"Data center administrators are constantly looking for ways to increase performance while simultaneously driving down costs, and one key component in that struggle is deploying the best JVM," said Guy Churchward, general manager, Java Runtime Products Group (JRPG), BEA Systems, Inc.
Intel has also had a hand in making JRockit perform well on their hardware.
"We have collaborated actively with BEA for years to ensure their software is optimized down to the JVM-level to take full advantage of our processor architectures," said Shannon Poulin, Enterprise Platform Marketing Director, Intel. "The result of this ongoing collaboration is a fast and robust infrastructure platform that is designed to allow IT departments to be more responsive to the needs of their businesses and provide an ideal foundation for service-oriented architectures."
9 replies so far (
Post your own)
Re: BEA JRockit 5.0 Sets New Performance Record
Sun's 1.5 JVM doesn't run on Itanium systems, so any claim that this result is faster than Sun's JVM on a machine "of the same configuration" is just plain stupid.Frankly, I'm glad Sun isn't expending the resources to optimize the JVM for a dead platform like the Itanium 2.
Re: BEA JRockit 5.0 Sets New Performance Record
You have to remember that BEA ultimately doesn't care. This is a project paid for by Intel development funds. They have to pretend that someone still uses Itanics and that the platform has a future.BEA in the mean time kept some engineer busy with others people money and now can say that they have their own really fast JVM.
Re: BEA JRockit 5.0 Sets New Performance Record
the last experience I'm using BEA JRockit was ends up with something like "outOfMemoryException". JRockit is a trash. Now everytime dealing with BEA Weblogic always using SUN JDK.Re: BEA JRockit 5.0 Sets New Performance Record
I remember JRockit being very fast with BEA. The problem was that it exposed some memory leaks in our application, only when we deployed it into our production environment. The VM handled object references differently from SUN's implemntation. I don't know about JRockit being trash just because of outOfMemoryExceptions, because Sun's VM will do that too. Does this mean Sun's VM is trash too?Re: BEA JRockit 5.0 Sets New Performance Record
Actually, Sun's 1.4.2 JVM supports Itanium so it could conceivably run on the same hardware.The press release is confusing though. A clearer statement would be "better than any 32-way score published using Sun's JVM" and/or "better than any score published on Sun hardware".
Re: BEA JRockit 5.0 Sets New Performance Record
Fast is good, but functional is better.JRockit 5.0 consistently core-dumped as soon as we tried to use Weblogic together with MQSeries. Needless to say, we switched to the Sun jvm which worked fine (this was on RedHat Linux, btw).
Anyone know if the JRockit problem is for JNI in general or if we encountered a "freak accident"?
Currently engaged in Weffo web architecture outline and the Flying Saucer xhtml and xml+css renderer
Re: BEA JRockit 5.0 Sets New Performance Record
From the JRockit release notes:"If you are using JRockit in conjunction with a native library that relies on OS signals you may experience crashes due to a signal handling conflict between JRockit and the native library.
Workaround: set the environment variable LD_PRELOAD as follows:
export LD_PRELOAD=$JROCKIT_HOME/jre/lib/i386/libjsig.so
BEA Engineering found this conflict using IBM's MQSeries native drivers, and it may be present in other libraries that rely on native code.
For more information, see http://java.sun.com/j2se/1.5.0/docs/guide/vm/signal-chaining.html "
Re: BEA JRockit 5.0 Sets New Performance Record
Thanks! Will try it ASAPCurrently engaged in Weffo web architecture outline and the Flying Saucer xhtml and xml+css renderer
Re: BEA JRockit 5.0 Sets New Performance Record
Thanks! Will try it ASAP.http://www.jvsoft.org