Comparing the Optimal Performance of Multiprocessor Architectures
Ansvarlig organisasjon
2003 (engelsk)Konferansepaper, Publicerat paper (Fagfellevurdert) Published
Abstract [en]
Consider a parallel program with n processes and a synchronization granularity z. Consider also two multiprocessors: a multiprocessor with q processors and run-time reallocation of processes to processors, and a multiprocessor with k processors and no run-time reallocation. There is an inter processor communication delay of t time units for the system with no run-time reallocation. In this paper we define a function g(n,k,q,t,z) such that the minimum completion time for all programs with n processes and a granularity z is at most g(n,k,q,t,z) times longer using the system with no reallocation and k processors compared to using the system with q processors and run-time reallocation. We assume optimal allocation and scheduling of processes to processors. The function g(n,k,q,t,z) is optimal in the sense that there is at least one program, with n processes and a granularity z, such that the ratio is exactly g(n,k,q,t,z). We also validate our results using measurements on distributed and multiprocessor Sun/Solaris environments.
sted, utgiver, år, opplag, sider
Innsbruck, Austria: Int. Assoc. of Science and Technology for Development, Calgery - Alberta, T3B OM6, Canada , 2003.
Emneord [en]
Multiprocessing systems, Computer architecture, Resource allocation, Scheduling, Synchronization, Computational complexity, Distributed system, Multiprocessors, Optimal bounds
HSV kategori
Identifikatorer
URN: urn:nbn:se:bth-9898Lokal ID: oai:bth.se:forskinfo7405B19EC3E39F26C1256DBF0026944CISBN: 0889863415 (tryckt)OAI: oai:DiVA.org:bth-9898DiVA, id: diva2:837877
Konferanse
21st IASTED International Multi-Conference Applied Informatics
2012-09-182003-10-142025-09-30bibliografisk kontrollert