by Tyng--ruey Chuang, Rong--guey Chang, Jenq Kuen Lee
http://falcon.cs.nthu.edu.tw/~jklee/papers/ppsc97.ps
Add To MetaCart
Abstract:
We present a compile--time method to select compression and distribution schemes for sparse matrices which are computed using Fortran 90 array intrinsic operations. The selection process samples input sparse matrices to determine their sparsity structures. It is also guided by cost functions of various sparse routines as measured from the target machine. The Fortran 90 array expression is then transformed into a sparse array expression that calls the selected compression and distribution routines. 1
Citations
|
87
|
Sparse matrices in Matlab: Design and implementation
– GILBERT, MOLER, et al.
- 1991
|
|
51
|
High-level optimization via automated statistical modeling
– Brewer
- 1995
|
|
38
|
Automatic data structure selection and transformation for sparse matrix computations
– Bik, Wijshoff
- 1996
|
|
25
|
A MATLAB to Fortran 90 translator and its effectiveness
– Rose, Padua
- 1996
|
|
19
|
New data-parallel language features for sparse matrix computations
– Ujaldon, Zapata, et al.
- 1995
|
|
17
|
Eicken, “LogP: A Practical Model of Parallel Computation
– Culler, Karp, et al.
- 1996
|
|
13
|
An experimental APL compiler for a distributed memory parallel machine
– Ching, Katz
- 1994
|
|
12
|
An array operation synthesis scheme to optimize Fortran 90 programs
– Hwang, Lee, et al.
- 1995
|
|
7
|
SPARSKIT: A basic tool kit for sparse computations, Version 2
– Saad
- 1994
|
|
5
|
A probabilistic approach to the problem of automatic selection of data representations
– Chuang, Hwang
- 1996
|
|
5
|
Issues in scalable library design for massively parallel computers
– Ni, Xu, et al.
- 1993
|
