|
ABSTRACT
Three approaches to parallelizing important components of the mathematical software package ELLPACK are considered: an explicit approach using compiler directives available only on the target machine, an automatic approach using an optimizing and parallelizing precompiler, and a two-level approach based on extensive use of a set of low level computational kernels. The focus is on shared memory architectures. Each approach to parallelization is described in detail, along with a discussion of the effort involved. Performance on a test problem, using up to sixteen processors of a Sequent Symmetry S81, is reported and discussed. Implications for the parallelization of a broad class of mathematical software are drawn.
REFERENCES
Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.
| |
1
|
Anderson , Bai , Bischof , Demmel , Dongarra , DuCroz , Greenbaum , Hammarling , McKenney , Sorenson, LAPACK Working Note 20: A Portable Linear Algebra Library For High-Performance Computers, University of Tennessee, Knoxville, TN, 1990
|
| |
2
|
|
| |
3
|
BONOMO, J P., DYKSEN, W. R, AND RICE, J.R. 1986. The ELLPACK performance evaluation system. Tech. Rep. CSD-TR 569, Dept. of Computer Sciences, Purdue Umv., West Lafayette, Ind.
|
| |
4
|
CHAKRABORTY, A., ALLISON, D. C. S., RIBBENS, C. J, ANn WATSON, L.T. 1989. Parallel orthogohal decompositions of rectangular matrices for curve tracking on a hypercube. In J. L Gustafson, Ed, Proceedings Fourth Conference on Hypercubes, Concurrent Computers, and Applications, ACM, New York, 651 654.
|
| |
5
|
CHAKRABORTY, A., ALLISON, D. C. S., RIEBENS, C. J., AND WATSON, L.T. 1991. Note on unit tangent vector computation for homotopy curve tracking on a hypercube. Parallel Camput. } 7, 1385 1395
|
| |
6
|
|
 |
7
|
|
| |
8
|
DONGARI~% J. J, BUNCH, J R., MOLER, C B., AND STEWART, G.W. 1979 LINPACK Users Guide. SIAM, Philadelphia, Pa.
|
 |
9
|
|
 |
10
|
|
| |
11
|
DuCRoz, J., MAYES, m., AND RADICATI, G. 1990. Factorizations of band matrices usmg level 3 blas. Tech. Rep. CS-90-109, Computer Scmnce Dept., Univ of Tennessee, Knoxville.
|
| |
12
|
DYKSEN, W. R., HOUSTIS, E. N., LYNCH, R. E , AND RICE, J.R. 1984. The performance of the collocatmn and galerkin methods with hermite bi-cubics SIAM d. Numer Anal. 21,695 715
|
| |
13
|
D'~SEN, W. R., RIBBENS, C. J., AND RICE, J. R 1988 The performance of numerical methods for elliptic problems with mixed boundary conditions Num. Meth PDEs 4, 347 361.
|
| |
14
|
HOUSTIS, E. N, RICE, J. R., CHRISOCHOIDES, N. m., KARATHANASIS, H. C., PAPACHIOU, P. N., SAMARTZIS, M. K, VAVALIS, E. A., AND WANG, K-Y. 1990. Parallel ELLPACK' A numerical simulation programming environment for parallel mired machines. Tech. Rep. CSD-TR-949, Dept. of Computer Sciences, Purdue Umv., West Lafayette, Ind.
|
| |
15
|
KINCAID, D. R., OPPE, T. C., AND YOUNG, D.M. 1989. ITPACKV 2D User's Guide. Tech. Rep. CNA-232, Center for Numerical Analysis, Univ. of Texas at Austin
|
 |
16
|
|
| |
17
|
KUCK & ASSOCIATES. 1989. KAP/Sequent User's Guide. Champaign, {1.
|
 |
18
|
|
| |
19
|
MAYES, P., AND RIDICATI, G. 1989 Banded cholesky factorization using level 3 bias. Tech. Rep. ANL/MCS-TM-134, Argonne National Lab., Argonne, I1.
|
| |
20
|
OPPE, T. C., AND KINCAID, D.R. 1990. Are there iterative blas? In Proceedings of the Copper Mountain Conference on Iterative Methods.
|
| |
21
|
|
| |
22
|
RAMDAS, M., AND K~NCAID, D.R. 1991. Parallelizing ITPACKV 2D for the Cray Y-MP. Tech. Rep. CNA-249, Center for Numerical Analysis, Univ. of Texas at Austin.
|
| |
23
|
|
| |
24
|
|
| |
25
|
RICE, J. R., HOUSTIS, E. N., AND DYKSEN, W.R. 1981. A population of linear, second order, elliptic partial differential equations on rectangular domains. Math. Comput. 36, 475-484.
|
| |
26
|
SKJELLUM, A., AND BALDWIN, C. 1991. The multicomputer toolbox: scalable parallel libraries for large-scale concurrent applications. Tech. Rep. UCRL-JC-109251, Lawrence Livermore National Lab.
|
| |
27
|
|
REVIEW
"Robert Charles Bell : Reviewer"
The problem of moving software to a parallel computing environment
is explored in this excellent paper. The authors take a model problem as
an example, using three important components of the ELLPACK mathematical
software package. Three approa
more...
|