Abstract
In the previous chapters, a large number of parallel machines and algorithms have been examined. Although there are exceptions, many of the algorithms studied can be placed into one of three categories. The first category consists of algorithms that communicate only between processors that are adjacent in a two-dimensional mesh. These algorithms will be called mesh-type algorithms. Of course, all plain mesh connected computer algorithms are mesh-type algorithms, but many algorithms for pyramid computers and hypercube computers with weak independent communication are in this class. As shown in Section 10.2, hypercube-derived computers can also implement mesh-type algorithms efficiently in an asymptotic sense, but these implementations do not appear to be practical. No technique is known for implementing mesh-type algorithms efficiently on a plain hypercube.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1994 Spinger-Verlag New York, Inc.
About this chapter
Cite this chapter
Cypher, R., Sanz, J.L.C. (1994). Conclusions. In: The SIMD Model of Parallel Computation. Springer, New York, NY. https://doi.org/10.1007/978-1-4612-2612-3_12
Download citation
DOI: https://doi.org/10.1007/978-1-4612-2612-3_12
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4612-7606-7
Online ISBN: 978-1-4612-2612-3
eBook Packages: Springer Book Archive