Abstract: "In this paper we consider problems that arise in a shared memory multiprocessor in which memory is physically distributed among a number of memories local to each processor or cluster of processors. The issue we address is that of deciding which local memories should contain copies o...

Ausführliche Beschreibung

Bibliographische Detailangaben

Hauptverfasser:	Black, David L. (VerfasserIn), Sleator, Daniel D. (VerfasserIn)
Format:	Buch
Sprache:	English
Veröffentlicht:	Pittsburgh, Pa. 1989
Schriftenreihe:	Carnegie-Mellon University <Pittsburgh, Pa.> / Computer Science Department: CMU-CS 89,201
Schlagworte:	Cache memory Multiprocessors
Zusammenfassung:	Abstract: "In this paper we consider problems that arise in a shared memory multiprocessor in which memory is physically distributed among a number of memories local to each processor or cluster of processors. The issue we address is that of deciding which local memories should contain copies of pages of data. In the migration problem we operate under the constraint that a page must be kept in exactly one local memory. In the replication problem we allow a page to be kept in any subset of the local memories, but do not allow a local memory to drop a page once it has it For interconnection topologies that are complete graphs, or trees we have obtained efficient on-line algorithms for these problems. Our migration algorithms also extend to interconnections that are products of these topologies (e.g. a hypercube is a product of simple trees). An on-line algorithm decides how to process each request (which is a read or write request from a processor to a page) without knowing future requests. Our algorithms are also said to be competitive because their performance is within a small constant factor of that of any other algorithm, including algorithms that make use of knowledge of future requests.
Beschreibung:	23 S.