Designing systems for highly parallel and distributed computing

Author

Vidovic, Nino ; Vrsalovic, Dalibor F. ; Budin, Leo

Author_Institution

Sch. of Comput. Sci., Carnegie Mellon Univ., Pittsburgh, PA, USA

Volume

i

fYear

1991

fDate

8-11 Jan 1991

Firstpage

75

Abstract

The parallel computational model with data flow sequencing is introduced. Description of the basic principles and features of mpDF, a massively parallel architecture, based on the dataflow operational model and RISC organizational principles is presented. The analytical performance model is developed to evaluate proposed architectural solution for distributed/parallel computing based on data flow sequencing of instructions. Algorithmic performance model is extended to include characterization of the parallel programs in terms of the average parallelism. Model is solved for number of different workloads. The values of the basic architectural characteristics are analyzed showing that architecture is well balanced providing consistent performance made for wide rage of parallel applications

Keywords

distributed processing; parallel architectures; performance evaluation; RISC; data flow sequencing; dataflow operational; distributed computing; highly parallel; massively parallel architecture; mpDF; parallel computational model; parallel programs; Analytical models; Computational modeling; Computer architecture; Concurrent computing; Data flow computing; Distributed computing; Parallel architectures; Parallel processing; Performance analysis; Reduced instruction set computing;

fLanguage

English

Publisher

ieee

Conference_Titel

System Sciences, 1991. Proceedings of the Twenty-Fourth Annual Hawaii International Conference on

Conference_Location

Kauai, HI

Type

conf

DOI

10.1109/HICSS.1991.183873

Filename

183873