Cilk vs MPI: comparing two very different parallel programming styles
作者： S. Tham , J. Morris
DOI: 10.1109/ICPP.2003.1240575
关键词:
摘要: We measured the relative performance of two support systems for parallel programming on networks workstations: Cilk - an extension C with dataflow semantics and MPI a commonly used library message passing. Although present significantly different processing models, we attempted to code benchmark problems in similar ways. The selected were matrix multiplication, travelling salesman problem, quick sort, Gaussian elimination, fast Fourier transform finite differencing. compared run times, speed-ups coding efficiency as by lines our implementations problems. showed speed advantage when smaller numbers large messages are transferred computation, enabling it gain more from underlying active implementation. algorithms natural solutions was compact, whereas which have simple iterative 'update-in-place' styles (Gaussian elimination differencing) efficiently expressed
sci-hub.se 参考文章(10)
Sonny Tham, John Morris, Performance of the Achilles Router Lecture Notes in Computer Science. pp. 365- 379 ,(2003) , 10.1007/978-3-540-39864-6_29
Mario Lauria, Scott Pakin, Andrew Chien, Efficient layering for high speed communication: the MPI over Fast Messages (FM) experience Cluster Computing. ,vol. 2, pp. 107- 116 ,(1999) , 10.1023/A:1019018423211
Jack Dongarra, Steven Huss-Lederman, David W. Walker, Steve W. Otto, Marc Snir, MPI: The Complete Reference ,(1996)
R.D. Blumofe, M. Frigo, C.F. Joerg, C.E. Leiserson, K.H. Randall, DAG-consistent distributed shared memory international conference on parallel processing. pp. 132- 141 ,(1996) , 10.1109/IPPS.1996.508049
William Gropp, Ewing Lusk, Nathan Doss, Anthony Skjellum, A high-performance, portable implementation of the MPI message passing interface standard parallel computing. ,vol. 22, pp. 789- 828 ,(1996) , 10.1016/0167-8191(96)00024-5
Robert D. Blumofe, Christopher F. Joerg, Bradley C. Kuszmaul, Charles E. Leiserson, Keith H. Randall, Yuli Zhou, Cilk: an efficient multithreaded runtime system acm sigplan symposium on principles and practice of parallel programming. ,vol. 30, pp. 207- 216 ,(1995) , 10.1145/209936.209958
David M. Young, Iterative Solution of Large Linear Systems ,(1971)
S. Tham, R. Gregg, John Morris, Achilles: high bandwidth, low latency, low overhead communication Achilles: high bandwidth, low latency, low overhead communication. pp. 173- 183 ,(1999)
William H. Press, Brian P. Flannery, Saul A. Teukolsky, William T. Vetterling, Harvey Gould, Numerical Recipes, The Art of Scientific Computing American Journal of Physics. ,vol. 55, pp. 90- 91 ,(1987) , 10.1119/1.14981
Saul A. Teukolsky, Brian P. Flannery, William T. Vetterling, William H. Press, Numerical recipes in C++ : the art of scientific computing Numerical recipes in C++ : the art of scientific computing by William H. Press. xxviii. ,(2002)