Modelling and analysis of the variance in parallelism in parallel computations

Sundaram, CRM and Narahari, Y (1993) Modelling and analysis of the variance in parallelism in parallel computations. In: Computers & Electrical Engineering, 19 (6). pp. 495-506.

PDF MODELLING-378.pdf Restricted to Registered users only Download (798kB) | Request a copy

Abstract

In this paper, we introduce an analytical technique based on queueing networks and Petrinets for making a performance analysis of dataflow computations when executed on the Manchester machine. This technique is also applicable for the analysis of parallel computations on multiprocessors. We characterize the parallelism in dataflow computations through a four-parameter characterization, namely, the minimum parallelism, the maximum parallelism, the average parallelism and the variance in parallelism. We observe through detailed investigation of our analytical models that the average parallelism is a good characterization of the dataflow computations only as long as the variance in parallelism is small. However, significant difference in performance measures will result when the variance in parallelism is comparable to or higher than the average parallelism.

Item Type:	Journal Article
Publication:	Computers & Electrical Engineering
Publisher:	Elsevier
Additional Information:	Copyright of thism article belongs to Elsevier.
Keywords:	Variance in parallelism;Average parallelism;Dataflow computations;Parallel computations;Product form queueing networks;Generalized stochastic Petri nets
Department/Centre:	Division of Electrical Sciences > Computer Science & Automation
Date Deposited:	26 Sep 2006
Last Modified:	19 Sep 2010 04:31
URI:	http://eprints.iisc.ac.in/id/eprint/8409

Actions (login required)

View Item