Efficient implementations of a class of ±2b parallel computations on a SIMD hypercube

Document Type

Conference Proceeding

Publication Date

1-1-1991

Abstract

We identify an important class of parallel computations, called ±2b - descend, with an efficient implementation on a hypercube. Given the input A[0: N - 1], a computation in this class consists of log N iterations. Iteration 6, 6 = logN - 1,.,0, computes the new value of each A[i] as a function of A[i], A[i+2b] and A[i-2b]. We obtain a general algorithm for implementing any computation in this class in O(logN) time on a SIMD hypercube. Our general descend algorithm results in an efficient O(log N) implementation of Batcher's odd-even merge algorithm on a hypercube. The best previously known implementation of odd-even merge on a SIMD hypercube requires O(log2 N) time.

Identifier

84943675164 (Scopus)

ISBN

[0818691670, 9780818691676]

Publication Title

Proceedings 5th International Parallel Processing Symposium IPPS 1991

External Full Text Location

https://doi.org/10.1109/IPPS.1991.153749

First Page

2

Last Page

9

This document is currently not available here.

Share

COinS