Performance tools on intel scalable high performance computing systems
A massive breakthrough of scalable high performance computing (SHPC) systems depends strongly on the ease of programming. Programming environments have to be provided wich enable the user to program and use SHPC systems in an efficient and transparent manner. A special challenge within these parallel programming environments are performance tuning tools. The paper gives an industrial perspective on the development of performance tools for SHPC systems with a focus on the ParAide programming environment for Intel's Paragon (tm) supercomputer. An integrated approach that provides an extensible and consistent programming environment for monitoring and visualizing the performance of parallel applications is presented. From the experiences made in the development of the Paragon performance toolset, future research directions are derived.
Unable to display preview. Download preview PDF.
- D. Breazeal, R. Anderson, W.D. Smith, W. Auld, and K. Callaghan. A Parallel Software Monitor for Debugging and Performance Tools on Distributed Memory Multicomputers. In Proceedings of the Supercomputer Debugging Workshop 1992, Dallas, Texas, October 1992. Los Alamos National Laboratory.Google Scholar
- D. Breazeal, K. Callaghan, and W.D. Smith. IPD: A Debugger for Parallel Heterogeneous Systems. In Proceedings of the ACM/ONR Workshop on Parallel and Distributed Debugging, pages 216–218, Santa Cruz, CA, May 1991.Google Scholar
- T. Bemmerl. The TOPSYS Architecture. In H. Burkhart, editor, CONPAR 90-VAPP IV, volume 457 of Lecture Notes in Computer Science, pages 732–743. Springer-Verlag, Berlin, Heidelberg, New York, September 1990.Google Scholar
- Intel Supercomputer Systems Division. Paragon XP/S Product Overview. Intel Corporation, 15201 N.W. Greenbrier Parkway, Beaverton OR 97006, 1992.Google Scholar
- S.L. Graham, P.B. Kessler, and M.K. McKusick. gprof: a Call Graph Execution Profiler. ACM SIGPLAN Notices, 17(6): 120–126, June 1982.Google Scholar
- M.T. Heath and J.A. Etheridge. Visualizing the Performance of Parallel Programs. IEEE Software, 8(5):29–39, September 1991.Google Scholar
- P. Pierce. The NX/2 Operating System. In Proceedings of the 3rd Conference on Hypercube Concurrent Computers and Applications, pages 384–391. ACM, 1988.Google Scholar
- B. Ries, R. Anderson, W. Auld, D. Breazeal, K. Callagham, E. Richards, W. Smith. The Paragon Performance Monitoring Environment. In Proceedings of Supercomputing '93, pages 850–859, Nov. 1993, Portland.Google Scholar
- D.T. Rover, M. B. Carter, and J. L. Gustafson. Performance Visualization of SLALOM. In Proceedings of the Sixth Distributed Memory Computing Conference, pages 543–550, Portland, Oregon, May 1991. IEEE, IEEE Computer Society Press.Google Scholar
- D. A. Reed, R. D. Olson, R. A. Aydt, T. M. Madhyastha, T. Birkett, D. W. Jensen, B. A. Nazief, and B. K. Totty. Scalable Performance Environments for Parallel Systems. In Proc. of the Sixth Distributed Memory Computing Conference, pages 562–569, Portland,Ore, April 1991. IEEE.Google Scholar
- C.Q. Yang and B.P. Miller. Critical Path Analysis for the Execution of Parallel and Distributed Programs. In Proceedings of the 8th International Conference on Distributed Computing Systems, pages 366–375, San Jose, CA, June 1988. IEEE.Google Scholar
- R. Zajcew, P. Roy, D. Black, C. Peak, P. Guedes, B. Kemp, J. LoVerso, M. Leibensperger, M. Barnett, F. Rabii and D. Netterwala. An OSF/1 UNIX for Massively Parallel Multicomputers. In Proceedings of the Winter 1993 USENIX Technical Conference, pages 449–468, San Diego, CA, January 1993. The USENIX Association.Google Scholar