Comments on: (edited) JavaOne talk on performance myths

By: Code Performance

Code Performance — Wed, 20 Jul 2005 21:47:56 +0000

Kirk, do you have any examples you can share with us?

http://java.sun.com/developer/community/chat/JavaLive/2004/jl1012.html
Guest: How do generics influence performance or are they performance neutral?

Scott Seligman: This is almost a follow-up to my previous answer. Since generics are dealt with (almost) entirely at compile-time only, there’s no performance impact, either better or worse.

By: Kirk

Kirk — Mon, 18 Jul 2005 06:29:59 +0000

To say that Auto-boxing is no more then syntactic sugar is the performance myth. It CAN affect performance once you move outside of the narrow range of choices that have been conocialized.

By: Code Performance

Code Performance — Thu, 14 Jul 2005 13:14:42 +0000

There was an interesting talk at javaone by Scott Oaks and Charlie Hunt on an nio based HTTP listener that Sun just opensourced in the glassfish project.

Regular I/O in Java is generally slow, and it’s gotten worse with every release that according to Sun, partially due to the JNI overhead. So try to use NIO, it may be complex but if performance and scalabilty are a concern that’s really the only way out.

How others have done it… http://www.manageability.org/blog/stuff/open-source-projects-that-use-java-nio/view

By: Rick Stabile

Rick Stabile — Tue, 12 Jul 2005 14:17:43 +0000

Well put. What do you have to say about what has, in the past, been a huge performance laggard for Java, namely File I/O?

By: Ann Onymous

Ann Onymous — Sun, 10 Jul 2005 15:00:04 +0000

It is very hard from the format you have this in to figure out what is a myth and what isn’t. Could you make this a little clearer? I realize this is probably just your notes, but if you could put it into sentences and put some kind of context around them, it would probably help a lot of people out.

Either that, or just link to the actual presentation.

Thanks!

By: Pablo F.

Pablo F. — Thu, 07 Jul 2005 07:35:22 +0000

nice summary! what about the comparison between ibm vm and hostpot vm for hastable performance, guess we’d have to wait for the slides…