SCALA: A framework for performance evaluation of scalable computing

  • Xian-He Sun
  • Mario Pantano
  • Thomas Fahringer
  • Zhaohua Zhan
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1586)


Conventional performance environments are based on profiling and event instrumentation. It becomes problematic as parallel systems scale to hundreds of nodes and beyond. A framework of developing an integrated performance modeling and prediction system, SCALability Analyzer (SCALA), is presented in this study. In contrast to existing performance tools, the program performance model generated by SCALA is based on scalability analysis. SCALA assumes the availability of modern compiler technology, adopts statistical methodologies, and has the support of browser interface. These technologies, together with a new approach of scalability analysis, enable SCALA to provide the user with a higher and more intuitive level of performance analysis. A prototype SCALA system has been implemented. Initial experimental results show that SCALA is unique in its ability of revealing the scaling properties of a computing system.


Execution Time Problem Size Scalability Analysis Graphical Object Range Comparison 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Adve, V. S., Crummey, J. M., Anderson, M., Kennedy, K., Wang, J.-C., and Reed, D. A. Integrating complication and performance analysis for data parallel programs. In Proc. of Workshop on Debugging and Performance Tuning for Parallel Computing Systems (Jan. 1996).Google Scholar
  2. 2.
    Benkner, S., Sanjari, K., Sipkova, V., and Velkov, B. Parallelizing ierregular applications with vienna HPF+compiler VFC. In HPCN Europe (April 1998), Lecture Notes in Computer Science, Springer-Verlag.Google Scholar
  3. 3.
    Calzarossa, M., Massari, L., Merlo, A., Pantano, M., and Tessera, D. Medea: A tool for workload characterization of parallel systems. IEEE Parallel & Distributed Technology Winter (1995), 72–80.Google Scholar
  4. 4.
    Calzarossa, M., Massari, L., Merlo, A., Pantano, M., and Tessera, D. Integration of a complication system and a performance tool: The HPF+ approach. In LNCS-HPCN98 (Amsterdam, NL, 1998).Google Scholar
  5. 5.
    Fahringer, T.Automatic Performance Prediction of Parallel Programs, Kluwer Academic Publishers, Boston, USA, ISBN 0-7923-9708-8, March 1996.zbMATHGoogle Scholar
  6. 6.
    Fox, G., Hiranandani, S., Kennedy, K., Koelbel, C., Kremer, U., Tseng, C., and Wu, M. Fortran D language specification. Technical Report, COMP TR90079, Department of Computer Science, Rice University, Mar. 1991.Google Scholar
  7. 7.
    Gustafson, J., Montry, G., and Benner, R. Development of parallel methods for a 1024-processor hypercube. SIAM J. of Sci. and Stat. Computing 9, 4 (July 1988), 609–638.zbMATHMathSciNetCrossRefGoogle Scholar
  8. 8.
    Hwang, K., and Xu, Z.Scalable Parallel Computing. McGraw-Hill WCB, 1998.Google Scholar
  9. 9.
    Kumar, V., Grama, A., Gupta, A., and Karypis, G.Introduction to Parallel Computing, Design and Analysis of Algorithms. The Benjamin/Cummings Publishing Company, Inc., 1994.Google Scholar
  10. 10.
    Lyon, G., Snelick, R., and Kacker, R. Synthetic-perturbation tuning of mimd programs. Journal of Supercomputing 8, 1 (1994), 5–8.zbMATHCrossRefGoogle Scholar
  11. 11.
    Noelle, M., Pantano, M., and Sun, X.-H. Communication overhead: Prediction and its influence on scalability. In Proc. the International Conference on Parallel and Distributed Processing Techniques and Applications (July 1998).Google Scholar
  12. 12.
    Reed, D., Aydt, R., Madhyastha, T., Noe, R., Shields, K., and Schwartz, B. An overview of the Pablo performance analysis environment. In Technical Report. UIUCCS, Nov. 1992.Google Scholar
  13. 13.
    Sahni, S., and Thanvantri, V. Performance metrics: Keeping the focus on runtime. IEEE Parallel & Distributed Technology (Spring 1996), 43–56.Google Scholar
  14. 14.
    Sun, X.-H. The relation of scalability and execution time. In Proc. of the International Parallel Processing Symposium’96 (April 1996).Google Scholar
  15. 15.
    Sun, X.-H. Scalability versus execution time in scalable systems. TR-97-003 (Revised May 1998), Louisiana State University, Department of Computer Science, 1997.Google Scholar
  16. 16.
    Sun, X.-H. Performance range comparison via crossing point analysis. In Lecture Notes in Computer Science, No 1388, J. Rolim, Ed. Springer, March 1998. Parallel and Distributed Processing.Google Scholar
  17. 17.
    Sun, X.-H., He, D., Cameron, K., and Luo, Y. A factorial performance evaluation for hierarchical memory systems. In Proc. of the IEEE Int’l Parallel Processing Symposium (Apr. 1999).Google Scholar
  18. 18.
    Sun, X.-H., Pantano, M., and Fahringer, T. Integrated range comparison for data-parallel compilation systems. IEEE Transactions on Parallel and Distributed Systems (accepted to appear, 1999).Google Scholar
  19. 19.
    Sun, X.-H., and Rover, D. Scalability of parallel algorithm-machine combinations. IEEE Transactions on Parallel and Distributed Systems (June 1994), 599–613.Google Scholar
  20. 20.
    SUN Microsystems Inc. Java 3D API specification., 1998.Google Scholar

Copyright information

© Springer-Verlag 1999

Authors and Affiliations

  • Xian-He Sun
    • 1
  • Mario Pantano
    • 2
  • Thomas Fahringer
    • 3
  • Zhaohua Zhan
    • 1
  1. 1.Department of Computer ScienceLouisiana State University
  2. 2.Department of Computer ScienceUniversity of IllinoisUrbana
  3. 3.Institute for Software Technology and Parallel SystemsUniversity of ViennaViennaAustria

Personalised recommendations