Java Performance Services
Training, Seminars, Benchmarking, Tuning

Java Performance Tuning Course


Chania Crete, May 17-20, 2010


Sun Extreme Learning EXL-2025

Houston, December 1-4,2009
New York, December 8-11, 2009
Washington DC, January 5-8, 2010



San Francisco, January 11-14

Anti-if

I have joined Anti-IF Campaign

Calendar

««Nov 2009»»
SMTWTFS
1234
5
67
891011121314
15161718192021
22232425262728
2930

Performance Anti-Patterns

My Top Tags

                                       

Mailing List

My RSS Feeds








JavaONE Day 2

posted Friday, 5 June 2009

Day 2 started off very good. I won a Acer Aspire netbook in an Intel./Sun talk on SPEC benchmarks. The SPEC talk was a great introduction into the various benchmarks offered by SPEC. It was also a good opportunity to talk about SPECJMS2007. I think that SPEC will be pulling the benchmark as it doesn’t measure what it is suppose to be measuring. Surprise surprise surprise, yet another broken bench. 


The nice news is that SPECJVM2008 is available as a free download. I’ve downloaded the bench but unfortunately it won’t run on Mac hardware as is. Well, I think there is a fix that will work with 1.6.0_07 but I’m now using 1.6.0_13 and that version seems to fix the problem that the scripts have been adjusted to handle. I’m sorting out how to back off “the fix” so hopefully I can get this thing to run.

The next talk I attended was on memory analysis using Eclipse MAT. Disappointing. IMHO, MAT is no where near as effective as NetBeans for finding memory leaks. Disappointing that the speakers didn’t know about jmap.

Next up, Crazy Bob on references. When we selected Bob’s talk the committee wondered how Bob was going to fill an hour. Well, he didn’t fill and hour and even in that time he filled in with a lot of (very interesting) Google stuff. The best part.. the wicked part was Bob’s demo of GC with reference types. It rocked! The amazing thing is it was.. well the whole presentation was written in Java. Kudo’s to Bob on a great talk.

 

Finally Alex Miller gave a solid talk on concurrency gotchas. The typical suspects showed up such as locking on non-final objects or objects that are too narrowly scoped. The slides from the talk are very informative and worth a scan