Journal of KIISE:Software and Applications (한국정보과학회논문지:소프트웨어및응용)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지

A Data Dependency Elimination Algorithm for Extracting Maximum Parallelism

최대 병렬성 추출을 위한 자료 종속성 제거 알고리즘

  • Published : 1999.01.01

⟨ Previous Next ⟩

Abstract

In most application programs, loops usually comprise most of the computation in a program and the most important source of parallelism. When the data dependency relation is uniformin terms of distance, several compile time parallelization methods were introduced. On the otherhand,when the data dependency relation is non-uniform in distance, the compile time extraction ofparallelism is much complicated. In this paper, a general method the extracting parallelism in nestedloops is presented. This algorithm can be applicable where the dependency relation is both uniform andnon-uniform in distance. According to execution repeatedly the statements in nested loops, thealgorithm which effectively removes these kind of data dependencies is developed in order to presentthe total parallelization of nested loops.

Keywords

References

  1. Proc.Int. Conf.Parallel Processing '84 The effect of program restructuring, algorithm changes, and architecture choice on program performance Kuck, D.J.;A.H.Smeh;R.Cytron;A.V.Veidenbaum (et al.)
  2. Comm.ACM v.29 no.12 Advanced compiler optimizations for supercomputers Padua, D.A.;M.Wolfe
  3. In 3rd Workshop on Languages and Compilers for parallel computing Unimodular transformations of double loops Banerjee, U.
  4. IEEE Parallel & Distributed Technology Defining, Analyzing and Transforming Program Constructs Li,J.;M.Wolfe
  5. IEEE Trans.on Parallel and Distributed Systems v.1 no.1 An Efficient Data Dependence Analysis for Paralleling Compilers Li,Z.;P.C.Yew;C.Q.Zhu
  6. IEEE Trans.on Parallel and Distributed Systems v.3 no.4 Partitioning and labeling of loops by unimodular transformations Hollander, E.H.
  7. Parallel Computing v.16 Optimal code parallelzation using unimodular transformations Dowling, M.L.
  8. Proceedings of International Conference Parallel Proceeding On loop transformations for generalized cycle shrinking Shang, W.;M.T.O'Keefe;J.A.B.Fortes
  9. The Journal of Supercomputing v.9 Beyond unimodular transfor-matjuons Ramanujam, J.
  10. IEEE Transaction on Parallel and Distributed Systems Loop Transformation Using Non-unimodular Matrices Agustin Fernandez;Jose M.Llaberia
  11. IEEE Trans. on Parallel and Distributed Systems v.4 no.5 Dependence Uniformization : A Loop Parallelzation Technique Tzen, T.H.;L.M.Ni
  12. In Journal of Parallel Algorithm and Architecture Compile time Partitioning of Nested Loop Iteration space with Non-uniform Dependencies Punyamurtula, V.;Chaudhary, J.Ju;S.Roy
  13. Unique Sets Orinted Partitioning of Nested Loops with Non-uniform Dependencies Ju., J.;V.Chaudhary
  14. Optimizing supercompiler for supercomputing Wolfe, M.E.
  15. Journal of parallel and distributed computing v.5 no.5 An overview of the PTRAN analysis system for multiprocessing Allen,F.;M.Burke;P.Charles;R.Cytron;J.Ferrante
  16. IEEE Trans.on Parallel and Distributed Systems v.3 no.5 The Power Test for Data Dependence Wolfe, M.E.;C.W.Tseng
  17. IEEE Trans.on Parallel and Sistributed Systems v.2 no.3 The I Text:An Improved Dependence Test for Automatic Parallelism and Vectorization Kong,X.;D.Klappholz;K.Psarris
  18. Proc.of the ACM International Conference on Supercomputing Compiler Techniques for Data Synchronization in Nested Parallel Loops Tang,P.;P.C.Yew;C.Q.Zhu
  19. More on Advanced Loop Optimization Polychronopoulos, C.D.
  20. IEEE Trans.on Comp. v.37 no.8 Compiler optimizations for enhancing parallelism and their impact on architecture design Polychronopoulos,C.D.
  21. Oregon Graduate Center 19600 NW won Newmann Drive Beaverton More Iteration Space Tiling Wolfe, M.J.
  22. Proc.of Intl.Conf.on Parallel Processing HAL:A High-Level Actor Language and Its Distributed Implementation Houck, C.;G.Agha
  23. IEEE, 1994 International Conference on Parallel Processing Parallel region execution of loops with irregular dependencies Zaafrani, A.;M.R.Ito
  24. ACM SIGPLAN Delinearization: An Efficient Way to Break Multiloop Dependence Equations Msalov, V.
  25. Proceedings of International Computer Software & Applications Conference Extracting Parallelism in Nested Loops 송월봉;박두순;김병수;공용해