Task Group Scheduling in Distributed Systems
Georgios L. Stavrinides, Helen D. Karatza · 2018
One of the main challenges in distributed systems is efficient scheduling of parallel applications. This paper addresses issues of scheduling parallel jobs that are bags-of-tasks on a cluster of distributed processors. It considers task group scheduling algorithms in two different cases of task routing. The goal is to examine the performance of the scheduling strategies in each of the task routing cases. Simulation is employed to evaluate the performance of the algorithms in different cases of system load. The simulation results reveal that the impact of the scheduling policies on the system performance depends on the employed routing policy, as well as the system load.