Comparative Study of Task Duplication Based Scheduling Algorithms for Parallel Systems

Scheduling of tasks is done by mapping tasks on multiple processors so that it requires least time for completion of all processes. Multiprocessors are used to run real time applications that uniprocessor systems would not be competent to execute. This paper presents various scheduling algorithms that schedule an edge-weighted Directed Acyclic Graph (DAG) to a number of processors. In this paper, task duplication based scheduling algorithms like PY algorithm and DSH algorithm are analyzed and studied for various performance matrices. Also, the effect of varying number of processors is examined on these algorithms.

Provided by: International Journal of Computer Applications Topic: Data Centers Date Added: Nov 2012 Format: PDF

Find By Topic