POEMS - End-to-End Performance Modeling of Large Heterogeneous Adaptive Parallel/Distributed Computer/Communication Systems

Purdue Group: Elias Houstis, John R. Rice, and Kihong Park

Collaborators: Jim Browne - University of Texas at Austin (Project Management), Vikram Adve - Rice University, Rajive Bagrodia - UCLA, Olaf Lubeck - Los Alamos Nat. Lab., Pat Teller - University of Texas at El Paso and Mary Vernon - University of Wisconsin

Support: DARPA

Project Description: The POEMS project will create and demonstrate a capability for prediction of the end-to-end performance of parallel/distributed implementations of large scale adaptive applications. The POEMS modeling capability will span applications, operating systems including parallel I/O, and architecture. Research will focus on the areas where there is little convention wisdom available in the execution behaviors of adaptive algorithms on multi-level memory hierarchies and in parallel I/O operations. POEMS will provide:

• a language for composing models from component models,
• derivation of models of applications as data flow graphs from HPF programs
• a library of component models spanning from workloads to memory hierarchies and at levels of resolution ranging large grain data flow graphs to instruction streams and from probabilistic to fully deterministic.
• parallel execution of the models
• a knowledge base of performance data for commonly used algorithms parameterized for architectural characteristics.

The POEMS development is driven by modeling a full-scale Department of Energy ASCI (Accelerated Scientific Computing Initiative) application code executing on an ASCI architecture. This version of POEMS focuses on high performance computational applications and architectures. The intent is that the POEMS technology can be applied to other large complex dynamic computer/communication systems such as GloMo and Quorum.

Recent Publications of the Purdue Group

1. Houstis, E.N., J.R. Rice, S. Weerawarana, A.C. Catlin, P. Papachiou, K.-Y. Wang, and M. Gastatzes, Parallel ELLPACK: A problem solving environment for PDE based applications on multicomputer platforms, ACM Trans. Math. Software, (1998), to appear.
2. Markus, S., S. Weerawarana, E.N. Houstis, and J.R. Rice, Scientific computing via the web: The net Pellpack PSE server, IEEE Comp. Sci. & Engr. 4 (No. 3), (1997), 43-51.
3. Houstis, E.N., S. Gallopoulos, R. Bramley, and J.R. Rice, Problem solving environments for computational science, IEEE Comp. Sci. & Engr. 4 (No. 3), (1997), 18-21.
4. Houstis, E.N., J.R. Rice, N. Ramakrishnan, T. Drashansky, S. Weerawarana, A. Joshi, and C.E. Houstis, Agent based systems to support multidisciplinary problem solving environments for computational science, to appear in Advances in Computers, (M. Zelkowitz, ed.), 1997. Also available as Purdue University Technical Report CSD-TR-97-031.
5. Joshi, A., N. Ramakrishnan, and E.N. Houstis, Multi-agent systems to support networked scientific computing, To appear in IEEE Internet Computing, (1997).
6. Ramakrishnan, N., A. Joshi, E.N. Houstis, and J.R. Rice, Collaborative environments for scientific computing: The task of algorithm/software selection, to appear in Mathematical Modelling and Scientific Computing, 8, 1997.
7. Ramakrishnan, N., E.N. Houstis, A. Joshi, J.R. Rice, and S. Weerawarana, Intelligent networked scientific computing, 15th IMACS World Congress, 4 Wissenschaft & Technik Verlag, Berlin, (1997), 285-790.
8. Drashansky, T.T., and J.R. Rice, SciAgents: solving complex heterogeneous PDE models using networks of interacting problem solvers, Math. Computing Modeling Sci. Comp., (1997)
9. Joshi, A., T. Drashansky, J.R. Rice, S. Weerawarana, and E.N. Houstis, Multi-agent simulation of complex hetergeneous models in scientific computing, Math. Computers Simulation, 44, (1997), 43-59.
10. Joshi, A., N. Ramakrishnan, T. Drashansky, E.N. Houstis, J.R. Rice, S. Weerawarana, and C.H. Tsoukalas, Agent based systems to support multidisciplinary problem solving environments, IEEE Comp. Sci. & Engr., (1997), to appear.
11. Drashansky, T.T., E.N. Houstis, A. Joshi, J.R. Rice, and S. Weerawarana, Multi-agent techniques for simulation of complex scientific systems, 15th IMACS World Congress, Wissenschaft & Technik Verlag, 4, (1997), 541-546.
12. Drashansky, T., A. Joshi, J. Rice, E. Houstis, and S. Weerawarana, A multi-agent environment for MPSEs, Parallel Proc. for Sci. Computing, SIAM Publications, http://www.siam.org/meetings/pp97/pp97home.htm, (1997), 8 pages.
13. Drashansky, T.T., A. Joshi, and J.R. Rice, Multidisciplinary problem solving using agents in a cluster environment, Cluster Computing Conf., http://www.mathcs.edu/~iccc97/sessions.html
14. Ramakrishnan, N., A. Joshi, E.N. Houstis, and J.R. Rice, Neuro-fuzzy approaches to collaborative scientific computing, Proc. Intl. Conf. Neural Networks, IEEE Press, (1997) to appear.
15. Weerawarana, S., E.N. Houstis, J.R. Rice, A. Joshi, and C. Houstis, PYTHIA: A knowledge based system to select scientific algorithms, ACM Trans. Math. Software, 22 (1996), 447-468.
16. Wu, P. and E.N. Houstis, Parallel adaptive mesh generation and decomposition, Engineering and Computers Journal, (1996).
17. Houstis, E.N., S-B. Kim, S. Markus, P.-T. Wu, C.E. Houstis, A.C. Catlin, and S. Weerawarana, Parallel ELLPACK elliptic PDE solvers, Proc. of 2nd Intl. Intel Supercomputer Users Group Conf., Albuquerque, New Mexico (1995).
18. Martin, I.M., D.C. Marinescu, and J.R. Rice, Adaptive load balancing strategies for solving irregular problems on distributed memory MIMD systems, Proc. 9th Intl. Par. Proc. Symp., IEEE Press, Los Alamitos, CA, (1995), 57-64.
19. Drashansky, T.T., A. Joshi, and J.R. Rice, SciAgents: An agent based environment for distributed, cooperative scientific computing, Proc. 7th Intl. Conf. Tools with Artificial Intelligence, IEEE Computer Society, Los Alamitos, CA, (1995), 452-459.
20. Ramakrishran, N., A. Joshi, S. Weerawarana, E.N. Houstis, and J.R. Rice, Neuro-fuzzy systems for intelligent scientific computation, in Intelligent Engineering through Artificial Neural Networks, Vol. 5: Fuzzy Logic and Evolutionary Programming , (C.H. Dagli et al., eds.), ASME Press, New York, (1995), 279-284.
21. N. Chrisochoides, E. Houstis, and J.R. Rice, Mapping algorithms and software environments for data parallel PDE iterative solvers, J. Par. Dist. Comp., 21 (1994), 75-95.
22. Houstis, E.N., A. Hadjidimos, S.B. Kim, and J.R. Rice, The performance of parallel stationary iterative methods for distributed memory machines, Proc. Intel Supercomputer User's group (D. Marinescu and R. Frost, eds) (1994), 169-173.
23. Marinescu, D.C., and J.R. Rice, On high level characterization of parallelism, J. Par. Dist. Comp., 20, (1994), 107-113.
24. Marinescu, D.C., and J.R. Rice, On the scalability of asynchronous parallel computations, J. Par. Dist. Comp., 22, (1994), 538-546.
25. Gallopoulos, E., E.N. Houstis, and J.R. Rice, Computer as thinker/doer: Problem solving environments for computational science, IEEE Comp. Sci. Engr., 1, (1994), 11-23.
26. Cornea-Hasegan, M.A., C. Costian, D.C. Marinescu, I. Martin, and J.R. Rice, Towards problem solving environments for high performance computing, High Performance Computing '94, National Supercomputer Research Center, Singapore, (1994), 354-366.
27. Marinescu, D.C., and J.R. Rice, Synchronization and load imbalance effects in distributed memory multi-processor systems, Concurrency: Practice and Experience, 3, (1991), 593-625.
28. Houstis, C.E., E.N. Houstis, J.R. Rice, S.M. Samartzis and D.L. Alexandrakis. The algorithm mapper: A system for modeling and evaluating parallel application/architecture pairs, In Intelligent Mathematical Software, (E.N. Houstis, J.R. Rice and R. Vichnevetsky, eds.) North-Holland, (1990), 87-101.
29. Houstis, C.E., E.N. Houstis, and J.R. Rice, Partitioning PDE computations: Methods and performance evaluation, J. Parallel Comp, 4 (1987), 143-163.
30. Boisvert, R.F., J. R. Rice, and E.N. Houstis, A system for performance evaluation of partial differential equations software, IEEE Trans. on Software Engr., 5 (1979), 418-425.