Dynamic task scheduling with precedence constraints and communication delays

Abstract

In this paper we have introduced the K1 heuristic algorithm for dynamic task scheduling with precedence constraints and communication delays. The execution of a task set repeats in cycles, while the execution and communication profile of a task set changes in time. During a task set execution, a new schedule is generated by tuning the previous schedule. The scheduling is distributed - performed on the processors of a highly parallel computer architecture. The goal is to reduce a dominant sequence of a task set during run-time. Only the tasks that can have an influence on dominant sequence reduction are considered for reordering/migration. The applied techniques are load balancing, task reordering, and data-wait reduction. Simulation is used for the algorithm validation and evaluation.