Low Overhead Parallel Schedules for Task Graphs

d Abstract) Richard J. Anderson Paul Beame Walter L. Ruzzo Department of Computer Science and Engineering University of Washingtoni We introduce a task scheduling model which is useful in the design and analysis of algorithms for small parallel machines. We prove that under our model, the overhead experienced in scheduling an n x n grid graph is O(loglogn) for p processors, p > 2. We also prove a matching lower bound of Q(loglog n) for p processors, p 1 2. We give an extension of the model to cover the case where the processors can have varying speed or are subject to delay.

• Type: Conference
• Year: 1990
• Where: SPAA
• Authors: Richard J. Anderson, Paul Beame, Walter L. Ruzzo
• Keywords: Beame Walter L, Distributed And Parallel Computing, Small Parallel Machines, SPAA 1990, Task Scheduling Model