Download:
|
by J. H. Lu, C. -h. Huang, R. W. Johnson
ftp://ftp.cis.ohio-state.edu/pub/hpce/tensor/Papers/HPCAsia95-sorting.ps.gz
Add To MetaCart
Abstract:
In this paper, we present the implementation of sorting algorithms based on bitonic mergers on vector multiprocessors. A programming methodology based on tensor products is used to design and implement these algorithms. We first demonstrate this methodology by expressing Batcher's bitonic merger in a tensor product formulation. Using the algebraic theory of tensor products, several variations of Batcher's bitonic merger are obtained. Programming characteristics of these formulas are described. An efficient sorting algorithm is constructed from these formulas. Finally, performance results for generated sorting codes on the Cray Y-MP are presented and compared with the Cray SciLib sorting routine.
Citations
|
1889
|
Matrix Analysis
– Horn, Johnson
- 1985
|
|
1446
|
The Art of Computer Programming
– Knuth
- 1973
|
|
380
|
Sorting networks and their applications
– Batcher
- 1968
|
|
173
|
Kronecker Products and Matrix Calculus: With Applications
– Graham
- 1981
|
|
152
|
Access and alignment of data in an array processor
– Lawrie
- 1975
|
|
72
|
On a class of multistage interconnection networks
– Wu, Feng
- 1980
|
|
54
|
A methodology for designing, modifying and implementing Fourier transform algorithms on various architectures
– Johnson, Johnson, et al.
- 1990
|
|
34
|
Radix sort for vector multiprocessors
– Zagha, Blelloch
|
|
24
|
Multilinear algebra and parallel programming
– Johnson, Huang, et al.
- 1991
|
|
20
|
A tensor product formulation of Strassen’s matrix multiplication algorithm with memory reduction
– Kumar, Huang, et al.
- 1995
|
|
12
|
An algebraic theory for modeling multistage interconnection networks
– Kaushik, Sharma, et al.
- 1993
|
|
5
|
Interconnections between processors and memory modules using shuffle-exchange networks
– Lang
- 1976
|
|
5
|
The reverse exchange interconnection network
– Wu, Feng
- 1980
|