I am very happy to know a person who has a common goal with me :-)
en, my research focus on how to get more useful information to guide compile thread and garbage collector to optimize the data locality and code quality, and my interest is how to use performance monitor to get profile , and i think few vitural machine use the performance monitor to guide code optimization automatic. Of course, many paper use PMU to guide the analysis. 2006/7/27, Mikhail Fursov <[EMAIL PROTECTED]>:
zouqiong, The dynamic optimization is an area I'm very interested in. I think I have enough of knowledge of how DRLVM DPGO subsystem works and can help you to implement or plug new profile collectors. What kind of dynamic optimizations you are going to research? Which kind of profile these optimizations require? We test the performance of drlvm with the benchmark specjbb2000, and the > score is a little lower than jdk1.5. I hope we will beat any other VM in future :). Right now I just want to remind testing of performance is not an easy task. There are a lot of dependencies on VM/JIT configuration settings. SpecJBB is one of the benchmarks that 1) depends on configuration heavily 2) cost a money. May be we can choose one of the free benchmarks for this local research? What about DeCapo hsqldb? -- Mikhail Fursov