On the use of meta-heuristics to increase the efficiency of online grid workflow scheduling algorithms

Abstract

The competitiveness of online algorithms is measured based on the correctness of the results produced and processing time efficiency. Traditionally evolutionary algorithms are not favored in online paradigms because of the large number of iterations involved in the algorithm which translates directly into processing time overhead. In this paper we describe MARS (Management Architecture for Resource Services) online scheduling algorithm which uses Simulated Annealing and concepts from Tabu Search to drastically decrease the processing time of the algorithm. The paper outlines the concepts behind MARS, the components involved and scheduling methodology used. In addition we also identify the time consuming bottlenecks in the performance of the system and how evolutionary algorithms help us soar past them.

Power efficient scheduling heuristics for energy conservation in computational grids

“…Priority Scheduling in MARS (PRISM) [10], the scheduling component that will form the basis of our discussion, quantifies a workstation based on how closely it matched to the requirements, and the Earliest Finish Time (EFT) of a task on the matched resources. One of the additional factors that PRISM considers while making the scheduling decisions is the possibility of allocating minimum number of workstations per job.…”

“…In our earlier work [10], we have discussed extensively how the composition of the computational grid directly affects the performance of the scheduler. Briefly, as the number of online workstations decrease in grid, the scheduling algorithm time improves.…”

Improving Energy-Efficiency of Scientific Computing Clusters

“…The approach does not require time characteristics of jobs being known. Aziz and El-Rewini (2008) studied online scheduling algorithms based on evolutionary algorithms in the grid context. Ni et al (2005) developed a heuristic scheduling algorithm for grid environments.…”