Автор |
Brunet, Jean-Philippe |
Автор |
Lennart Johnsson, S. |
Дата выпуска |
1992 |
dc.description |
An all-to-all broadcast algorithm that exploits concur rent communication on all channels of the Connection Machine system CM-200 binary cube network is de scribed. Issues in integrating a physical all-to-all broad cast between processing nodes into a language envi ronment using a global address space are discussed. Timings for the physical broadcast between nodes and for the virtual broadcast are given. The peak data transfer rate for the physical broadcast on a CM-200 is 5.9 gigabytes/sec, and the peak rate for the virtual broadcast is 31 gigabytes/sec. Array reshaping is an effective performance optimization technique. An ex ample is given where reshaping improved perfor mance by a factor of 7 by reducing the amount of local data motion. We also show how to exploit symmetry for computation of an interaction matrix using the all- to-all broadcast function. Further optimizations are suggested for N-body-type calculations. Using the all- to-all broadcast function, a peak rate of 9.3 GFLOPS/ sec has been achieved for the N-body computations in 32-bit precision on a 2,048 node Connection Machine system CM-200. |
Издатель |
Sage Publications |
Название |
All-To-All Broadcast and Applications On the Connection Machine |
Тип |
Journal Article |
DOI |
10.1177/109434209200600303 |
Print ISSN |
1094-3420 |
Журнал |
International Journal of High Performance Computing Applications |
Том |
6 |
Первая страница |
241 |
Последняя страница |
256 |
Аффилиация |
Brunet, Jean-Philippe, THINKING MACHINES CORPORATION AND HARVARD UNIVERSITY CAMBRIDGE, MASSACHUSETTS |
Аффилиация |
Lennart Johnsson, S., THINKING MACHINES CORPORATION AND HARVARD UNIVERSITY CAMBRIDGE, MASSACHUSETTS |
Выпуск |
3 |
Библиографическая ссылка |
Applegate, J.H., Douglas, M.R., Gursel, Y., Hunter, P., Seitz, C.L., and Sussman, G.J.1985. A digital orrery. IEEE Trans. Compul.34(9):822-831. |
Библиографическая ссылка |
Bertsekas, D. P., Ozveren, C., Stamoulis, G. D., Tseng, P., and Tsitsiklis, |
Библиографическая ссылка |
J N.1991. Optimal communication algorithms for hypercubes. J. Parallel Distributed Comput.11: 263-275. |
Библиографическая ссылка |
Brunet, J.-P., Mesirov, J.P., and Edelman, A.1990. An optimal hypercube direct N-body solver on the Connection Machine. In Supercomputing 90. Los Alamitos, Calif.: IEEE Computer Society Press, pp. 748-752. |
Библиографическая ссылка |
Fox, G.C., and Furmanski. W.1986. Optimal communication algorithms on hypercube. Technical Report CCCP-314Pasadena: California Institute of Technology. |
Библиографическая ссылка |
Hennessy, J.L., and Patterson, D.A.1990. Computer architecture: a quantitative approach . San Mateo, Calif.: Morgan Kaufmann Publishers. |
Библиографическая ссылка |
Johnsson, S.L., and Brunet, J.-P.1991. Exploiting symmetry in computing interaction matrices. Technical ReportCambridge, Mass.: Thinking Machines Corp. |
Библиографическая ссылка |
Johnsson, S.L., and Ho, C.-T.1992. Generalized shuffle permutations on Boolean cubes . J. Parallel Distributed Comput. 16(1):1-14. |
Библиографическая ссылка |
Johnsson, S.L., and Ho, C.-T.1989. Spanning graphs for optimum broadcasting and personalized communication in hypercubes. IEEE Trans. Comput.38(9): 1249-1268. |
Библиографическая ссылка |
Johnsson, S.L., and Mathur, K.K.1991. Distributed BLAS. Technical ReportCambridge, Mass.: Thinking Machines Corp. |
Библиографическая ссылка |
Mathur, K.K., and Johnsson, S.L.1992. All-to-all communication. Technical Report 243. |
Библиографическая ссылка |
Cambridge, Mass.: Thinking Machines Corp. |
Библиографическая ссылка |
Reingold, E.M., Nievergelt, J., and Deo, N.1977. Combinatorial algorithms. Englewood Cliffs. N.J.: Prentice-Hall. |
Библиографическая ссылка |
Stout, Q.F., Wagar, B.1987. Passing messages in link-bound hypercubes. In Hypercube multiprocessors 1987, edited by M. T. Heath.Philadelphia: Society for Industrial and Applied Mathematics. |
Библиографическая ссылка |
Thinking Machines Corp.1991a. CM Fortran Optimization Notes: Slicewise Model, Version 1.0. |
Библиографическая ссылка |
Thinking Machines Corp. 1991b. CM-Fortran Reference Manual. |