Parallel and Distributed Processing Symposium, International (2003)
Apr. 22, 2003 to Apr. 26, 2003
Guang R. Gao , University of Delaware
Kevin B. Theobald , University of Delaware
R. Govindarajan , University of Delaware
Clement Leung , University of Delaware
Ziang Hu , University of Delaware
Haiping Wu , University of Delaware
Jizhu Lu , University of Delaware
Juan del Cuvillo , University of Delaware
Adeline Jacquet , University of Delaware
Vincent Janot , University of Delaware
Thomas L. Sterling , California Institute of Technology
Future high-end computers which promise very high performance require sophisticated program execution models and languages in order to deal with very high latencies across the memory hierarchy and to exploit massive parallelism. This paper presents our progress in an ongoing research toward this goal. Specifically, we will develop a suitable program execution model, a high-level programming notation which shields the application developer from the complexities of the architecture, and a compiler and runtime system based on the underlying models. In particular, we propose fine-grain multithreading and thread percolation as key components of our program execution model. We investigate implementing these models and systems on novel architectures such as the HTMT architecture and IBM?s Blue Gene. Also, we report early performance prediction of thread percolation and its impact on execution time.
T. L. Sterling et al., "Programming Models and System Software for Future High-End Computing Systems: Work-in-Progress," Parallel and Distributed Processing Symposium, International(IPDPS), Nice, France, 2003, pp. 206b.