Scheduling Single-Task Jobs along with Bag-of-Task-Chains in Distributed Systems

Abstract

Distributed systems need to be used efficiently in order to achieve good performance. Particularly important in these systems are resource allocation and scheduling. In this paper we study the scheduling of complex workloads that include regular single-task jobs as well as bag-of-task-chains. Two algorithms are examined for the scheduling of the workload in a distributed system. Simulation is employed to evaluate their performance for various workload conditions. The results reveal that the performance of the scheduling techniques depends on the type of the workload and the system load.