Workshop on Randomized Parallel Computing Panos Pardalos, University of Florida, Gainesville Sanguthevar Rajasekaran, University of Florida, Gainesville

Parallel and Distributed Processing

Volume 1388 of the series Lecture Notes in Computer Science pp 418-427

Date:

Implementing parallelism in random discrete event-driven simulation

  • Marc BumbleAffiliated withThe Pennsylvania State University
  • , Lee CoraorAffiliated withThe Pennsylvania State University

* Final gross prices may vary according to local VAT.

Get Access

Abstract

The inherently sequential nature of random discrete event-driven simulation has made parallel and distributed processing difficult. This paper presents a method of applying Reconfigurable Logic to gain some parallelism in event generation. Field Programmable Gate Arrays (FPGAs) are used to create a flexible and fast environment in which events may be generated according to various statistical models. The method presented accelerates both event generation and the elimination of some blocked events from the Event Queue.