Meta-Heuristically Seeded Genetic Algorithm for Independent Job Scheduling in Grid Computing

Abstract

Grid computing is an infrastructure which connects geographically distributed computers owned by various organizations allowing their resources, such as computational power and storage capabilities, to be shared, selected, and aggregated. Job scheduling problem is one of the most difficult tasks in grid computing systems. To solve this problem efficiently, new methods are required. In this paper, a seeded genetic algorithm is proposed which uses a meta-heuristic algorithm to generate its initial population. To evaluate the performance of the proposed method in terms of minimizing the makespan, the Expected Time to Compute (ETC) simulation model is used to carry out a number of experiments. The results show that the proposed algorithm performs better than other selected techniques.

A Loosely Coupled Hybrid Meta-Heuristic Algorithm for the Static Independent Task Scheduling Problem in Grid Computing

“…ACO seems an appropriate candidate for the problem of task scheduling in computational grids as it has been successfully used for various NP complete optimization problems. This section introduces a loosely coupled hybrid ACO+VNS algorithm for the problem of task scheduling in grid systems, which uses the modified ACO described in our previous work [10].…”

“…It is worth noting that we used different rules for updating the pheromone trail and for ∆τ tm in our previous work [10].…”

“…In our previous work [10], a loosely coupled hybrid algorithm which combined a newly proposed ACO and the GA proposed in [11], i.e., ACO+GA, was suggested as a promising algorithm for solving the task scheduling problem in grid computing. The ACO starts first, the output from which will be used by the GA proposed in [11], which further improves it.…”

“…The proposed VNS is hybridized with a meta-heuristic method in a loosely coupled fashion, yielding a new hybrid meta-heuristic algorithm by which to consider the task scheduling problem in grid computing. The new algorithm, called ACO+VNS, combines a modified ACO from our previous work [10] and VNS, in which the former works first and whose output is further refined by the latter algorithm. To evaluate the proposed method in terms of minimizing the makespan, the ETC simulation model is used to carry out a number of experiments.…”

“…Algorithm 3 demonstrates the general steps of the proposed hybrid ACO+VNS algorithm. It is worth noting here that in our previous work [10], a small value was used to initialize the pheromone deposit only. In this study, Eq.…”