Slashdot Mirror

Second: profile, profile, profile

Third: well, almost anybody that has used a J2SDK (or JRE) on Solaris knows about its problems. Try to run Volano's benchmark to know more about this. But like any banchmark, please don't believe your software will perform the same way the benchmark does. It is just an indicative.

There is a memo about this problem, supposedly from Sun. If the problem realy exists (I know it does, but you should find it by yourself), you'll know your Solaris servers will not deliver as much transactions as other power processing equivalent servers.

If your concearns are all about costs, you should make tests with x86 solutions. Some big players like IBM and HP will let you make some tests on a test machine (specially if your transition is successful and you let them put your case in an add ;-)

Second: profile, profile, profile

Third: well, almost anybody that has used a J2SDK (or JRE) on Solaris knows about its problems. Try to run Volano's benchmark to know more about this. But like any banchmark, please don't believe your software will perform the same way the benchmark does. It is just an indicative.

There is a memo about this problem, supposedly from Sun. If the problem realy exists (I know it does, but you should find it by yourself), you'll know your Solaris servers will not deliver as much transactions as other power processing equivalent servers.

If your concearns are all about costs, you should make tests with x86 solutions. Some big players like IBM and HP will let you make some tests on a test machine (specially if your transition is successful and you let them put your case in an add ;-)

This is not exactly an accurate test. In many cases the author states that since the "stock" version of the feature runs slow in C/++, he has provided a replacement.

Any truly accurate test would have to ignore most of the "stock" implementations in all languages. The composer of the study would also have to be equally skilled in the languages, for the test to be accurate.

It would also be good to know which JVM he was using to test with. I assume it is the one he used to compile with (JDKSE 1.4.1-02 by Sun). The Sun reference versions have been shown in other tests to be inferior to implementations by other vendors. As we can see here, different implementations have VERY different performances.

To sum up, while this may be a fair comparison between C# and the C derivatives, it does not seem to be an objective study of ALL of the langauges tested.

Sorry but performers of this benchmark can't be that ignorant of the performance issues of the Sun's JVM on Linux. Check out this (somewhat outdated) report to find out that other JVM's (including IBM's JVM which is free as in beer) are much faster than Sun in performing most tasks.

Osvaldo Pinali Doederlein.
The Java Performance Report - Part IV: Static Compilers, and More. August, 2001

The Java Performance Report is an independent study of Java performance, where both virtual machines and static compilers are evaluated. Part IV compares Excelsior JET and two other static compilers with Sun's HotSpot Server VM.

Volano, LLC.
The Volano Report. Dec 26, 2001

The Volano Report covers VolanoMark(tm) - a widely recognized pure Java server benchmark characterized by long-lasting network connections and high thread counts. This report includes Excelsior JET a native compiler.

Excelsior JET Benchmarks

From Caffeine and SPECJVM through XML Transformations to Java2Demo and VolanoMark.

Starting with JDK1.4 the releases will be synchronized. On a side note, there are blazing fast JVMs (no pun intended) for wintendo, check out IBM's VM on http://www.volano.com/report.html.

The really sad part about the resistance to Java is that the excellent work of the Blackdown Project, Kaffe and projects like Java Gnome seem understaffed, yet Linux is one of the best platforms for running Java. Linux as a platform beats out most other platforms (like Solaris and NT), as seen in the latest Volano Report. And Java as a language keeps getting faster vs. C all the time. If Sun would finally realize Java is bigger than they are, Java would finally get the recognition it deserves.

And in closing, Java on Linux is dope.

F.O. Dobbs
Portal-Potty Founder and Mr. Brown drinker.

• SPEC JBB2000
• Volano
• Java Grande Forum - not a benchmark, but these folks work on improving Java for high-performance computing

In order to help you choosing the best VM, you can check the Volano Benchmark .
You will see there that the best VM is Tower TowerJ 3.1.4 for Linux !
Second point, I never doubt that java on the server is a good solution now. For me, the only trouble with java now is the memory gluttony.
If some of you want to test Jsp/Servlet, here are some good open source products : java.apache.org (JSP, servlet), www.enhydra.org (JSP, servlet, EJB)

Just ran some benchmarks, looks like some solid progress on the IBM JVM under threadload. IBM's JVM is now runs just short of TowerJ's performance.
If you're not familure with TowerJ, it's a commerical piece of software which given a set of java class files, turns them into C++ code. Then runs GCC over them, producing a native binary. TowerJ [specifically on Linux] is currently ranked in the top slot on the Volano Benchmark. TowerJ under linux even beats TowerJ under NT; apparently in the words of the TowerJ engineers, the linux libraries are just more efficient :)

You can view a JVM graph here of the various Linux JVMs under load. As you can see the old IBM JVM didn't fair very well under threadload, regardless that it generally ran faster for single threaded applications. Good work IBM!

It's worth having a look at what your java code is doing - and select your implementation of it with care. http://ww w.volano.com/report.html</A> will give you a good comparison of a few of the brands out there. I was working on a project using a java front end to an informix db at my last job, and one problem that I vaguely recollect was something like all reads on hashtables being synchronous, i.e. only one thread could read from it at a time. Those sorts of things can produce severe bottlenecks. I think IBM provide some fairly good runtime analysis tools with their jikes compiler for free, have a look around their <A href="http://www10.software.ibm.com/developer/open source/jikes/project/index.html">site</A >.

Juergen

Check out the Volano Report. Sun/Solaris and Blackdown/Linux rock....

The 1.1.8 runtime does have a speed advantage. You should see its performance on NT and OS/2. It's faster than anyone's 1.2 implementation.

http://www.volano.com/report.html

What seems to be very promising for Linux is IBM's JDK once they fix the scaleability issues, which I have a lot of confidence they will. Why? They seem to be undergoing a transformation recently and producing a boatload of great stuff, especially for Linux.

IBM, thanks for supporting Linux with a speedy JIT-enabled VM (see The latest Volano Report). But you could help out the entire Java world by licensing/GPL'ing the JIT for other platforms?

Just thinking out loud: Wouldn't it be great if you could get from IBM a GPL'ed VM and JIT for all known platforms? Instead of every platform rolling their own port, they could check out the VM code, make changes for their platform, and then check it back in. This is what, IMO, Java should have been - all one source - both VM and JIT.

Then Java would really run everywhere and it would run fast!

Nobody beats linux for server-side java
Not even Sun. This, of course, is on x86 hardware. The larger boxes (AIX, sparc, alpha) blow these numbers off the charts.

javaworld recently did a decent comparison of jvms. I would post the link, but their site is down. To summarize, the notion that the win32 jvm/jit combos are best is passe. blackdown is moving up, and will surpass the ms and sun jvms on win32. kaffe is not reviewed (ifirc), but it smoked blackdown on the volano benchmarks.

It sounds like you should be using Java on OS/2. According to the Volano benchmark, the OS/2 JVM was beaten only by the TowerJ compiler on Linux.

Well, for java/os speed you should check this out..
www.volano.com/report.html