Principal Investigator

• James Browne
  Department of Computer Science
  University of Texas
  Austin, Texas 78712
  Phone: 512-471-9579, FAX: 512-471-8885
  browne@cs.utexas.edu
  

Co-Principal Investigators

• Vikram Adve
  Rice University, CRPC - MS 41
  6100 South Main Street
  Houston, TX 77005-1892
  Phone: 713-285-5934, Fax: 713-285-5136
  adve@rice.edu
  
  
• Rajive Bagodia
  Computer Science
  University of California at Los Angeles
  Los Angeles, California 90095
  Phone: 310-825-0956, Fax: 310-825-2273
  rajive@cs.ucla.edu
  
  
• Elias Houstis
  Computer Science
  Purdue University
  West Layfayette, Indiana 47907
  Phone: 765-494-6181, Fax: 765-494-0739
  enh@cs.purdue.edu
  
  
• John Rice
  Computer Science
  Purdue University
  W. Lafayette, IN 47907
  Phone: 765-494-6007, Fax: 765-494-0739
  rice@cs.purdue.edu
  
  
• Olaf Lubeck
  Scientific Computing Group
  Los Alamos National Laboratory
  Los Alamos, New Mexico 87454
  Phone: 505-667-6017, Fax: 505-667-1126
  oml@lanl.gov
  
  
• Patricia Teller
  Computer Science
  University of Texas at El Paso
  El Paso, Texas 79968-0518
  Phone: 915-747-5939, Fax: 915-747-5030
  pteller@cs.utep.edu
  
  
• Mary Vernon
  Computer Science Department
  1210 W. Dayton Street
  Madison, WI 53706-1685
  Phone: 608-262-7893, Fax: 608-262-9777
  vernon@cs.wisc.edu
  
  

OBJECTIVE

The goals of this project are:
1. to develop a comprehensive environment (POEMS) for end-to-end performance analysis of large, heterogeneous adaptive parallel/distributed computer and communication systems, and
2. to demonstrate the use of the environment in analyzing and improving the performance of defense-critical scalable systems.

APPROACH

Second, we will experimentally and incrementally develop and validate scalable models. This will involve using multi-scale models, multi-paradigm models, and parallel model execution in complementary ways. Multi- scale models will allow different components of a system to be modeled at varying levels of detail via the use of adaptive module interfaces, supported by the specification language. Multi-paradigm models will allow an analyst to use the modeling paradigm -- analytical, simulation, or the software or hardware system itself -- that is most appropriate with respect to the goals of the performance study. Integration of an associative model of communication with data mediation methods to provide adaptive component interfaces will allow us to compose disparate models in a common modeling framework. To handle computationally expensive simulations of critical subsystems in a complex system, we will incorporate parallel simulation technology based on the Maisie language.

Third, we will provide a library of models, at multiple levels of granularity, for modeling scalable systems like those envisaged under the DOE ASCI program, and for modeling complex adaptive systems like those envisaged under the GloMo and Quorum programs.

Finally, we will provide a knowledge base of performance data that can be used to predict the performance properties of standard algorithms as a function of architectural characteristics.

CURRENT PLAN

A significant segment of the component library models will be developed and validated in FY 1998. In particular the first versions of models for memory hierarchies, for parallel I/O and for high performance networks will be developed and validated in FY 1998.

The first end-to-end models of relatively simple applications on homogeneous parallel computers will be developed and validated in FY 1998.

The initial version of the knowledge base on the performance of commonly used algorithms on existing parallel architectures will be established in FY 1998.

The initial implementation of the model development environment utilizing the component models and the knowledge base and compiling the specification language will become available for use in FY 1998.

Return to: POEMS homepage