Multiagent and Grid Systems – An International Journal 11 (2015) 227–244 227 DOI 10.3233/MGS-150237 IOS Press A hierarchical approach for job scheduling in grid computing based on resource prediction and meta-heuristic algorithms Elahe Goudarzi and Reza Ravanmehr * Computer Engineering Department, Central Tehran Branch, Islamic Azad University, Tehran, Iran Abstract. The computational grids as a distributed system are hardware and software infrastructures that are capable of solving large-scale issues, and they use heterogeneous or homogeneous resources scattered around the globe by a high-speed network. Scheduling is a critical and prominent issue in grid computing. An appropriate prediction method for allocating jobs to resources may significantly affect quality of service parameters. In this paper, a hierarchical approach is presented for job scheduling in computational grid utilizing a resource prediction method based on the scoring system. It is inspired by meta-heuristic algorithms in order to improve parameters such as makespan, load balancing and the rate of meeting deadlines. To evaluate the proposed method, GridSim toolkit is exploited. According to the simulation results and comparison with some recent well- known methods, this approach has been successful in improving the mentioned parameters. Keywords: Computational grid, scheduling, quality of service, heuristic algorithm, GridSim 1. Introduction The fast growing of different scientific domains and complex computations has led to increasing re- quirement of computational power. In the past, expensive super computers were utilized to support these computations; whereas, nowadays computational grids could be employed. They provide capabilities similar to supercomputers while reducing the costs of computation considerably [19]. Computational grid is a geographically distributed system consisting of heterogeneous resources. Since it aims to perform large scale complicated computations and resources are common between users, a critical issue in computational grid is proper allocation of tasks to resources in order to reduce makespan. Thus, resource management and job allocation, or in the other words an efficient scheduling is crucial in such systems. One of the scheduling requirements is appropriate architecture, scheduling architectures in the grid might be centralized, decentralized or hierarchical [9]. In centralized architecture, a global scheduler is connected to all resources directly. This type of architecture is utilized by many scheduling algorithms such as traditional algorithms [19]. In decentralized architecture, each resource has its own scheduler. It is used in methods where regardless of cost the fault tolerance is important [19]. Hierarchical archi- tecture is a combination of centralized and decentralized that includes a global scheduler and several * Corresponding author: Reza Ravanmehr, Computer Engineering Department, Central Tehran Branch, Islamic Azad Univer- sity, Tehran, Iran. E-mail: r.ravanmehr@iauctb.ac.ir. ISSN 1574-1702/15/$35.00 c  2015 – IOS Press and the authors. All rights reserved