Abstract: Programming languages that are executed by virtual machines face significant performance challenges beyond those confronted by traditional languages. First, portable program representations and dynamic language features, force the deferral of most optimizations until runtime, inducing runtime optimization overhead. Second, modular program representations preclude many forms of whole-program interprocedural optimization. Third, virtual machines incur additional costs for runtime services such as security guarantees and automatic memory management. To address these challenges, mainstream virtual machine implementations include substantial infrastructure for online profiling, dynamic compilation, and feedback-directed optimization. This talk will survey the state-of-the-art in the areas of dynamic compilation and adaptive optimization in virtual machines by debunking several misconceptions about these two topics.

This talk is based on PLDI'04 tutorial created by Stephen Fink, David Grove, and Michael Hind. The complete slides from this 3-hr tutorial are available at Michael Hind's web page.

Short Bio: Michael Hind received his Ph.D. degree from New York University in 1991. From 1992-1998, Michael was an assistant and associate professor of computer science at the State University of New York at New Paltz. In 1998, Michael became a Research Staff Member in the Software Technology Department at the IBM T.J. Watson Research Center, working on the Jalapeno project, the project that produced the open source Jikes RVM. In 2000, he became the manager of the Dynamic Optimization Group at IBM Research. Michael is a member of the Jikes RVM steering committee and core teams. His research interests include adaptive optimization, program analysis, and software optimizations to address memory latency.