Methods for Message Routing in Parallel Machines
Tom Leighton
Additional contact information
Tom Leighton: Massachusetts Institute of Technology, Mathematics Department and Laboratory for Computer Science
A chapter in Proceedings of the International Congress of Mathematicians, 1995, pp 1417-1417 from Springer
Abstract:
Abstract The problem of getting the right data to the right place within a reasonable amount of time is one of the most challenging and important tasks facing the designer (and, in some cases, the user) of a large-scale general-purpose parallel machine. This is because the processors comprising a parallel machine need to communicate with each other (or with a common shared memory) in a tightly constrained fashion in order to solve most problems of interest in a timely fashion. Supporting this communication is often an expensive task, in terms of both hardware and time. In fact, most parallel machines devote a significant portion of their resources to handling communication between the processors and the memory. In the talk, we surveyed several of the ideas and approaches that have been proposed for solving communication problems in parallel machines. Particular emphasis was placed on recent work involving randomly wired networks (known as multibutterflies). Results were presented that indicate that randomly wired networks significantly outperform traditional networks such as the butterfly in terms of both speed and fault tolerance.
Date: 1995
References: Add references at CitEc
Citations:
There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.
Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
HTML/Text
Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-0348-9078-6_136
Ordering information: This item can be ordered from
http://www.springer.com/9783034890786
DOI: 10.1007/978-3-0348-9078-6_136
Access Statistics for this chapter
More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().