A Novel Genetic Algorithm for Effective Job Scheduling in Grid Environment
A grid is a set of resources such as CPU, memory, disk, applications, and database distributed over wide area networks and supports large-scale distributed applications. Resources in grid are geographically distributed and linked through Internet, to crea
- PDF / 1,103,991 Bytes
- 9 Pages / 439.37 x 666.142 pts Page_size
- 14 Downloads / 183 Views
Abstract A grid is a set of resources such as CPU, memory, disk, applications, and database distributed over wide area networks and supports large-scale distributed applications. Resources in grid are geographically distributed and linked through Internet, to create virtual supercomputer with vast computing capacity to solve complex problems. Scheduling, resource brokering, and load balancing are the essential functionalities of grid environment. Evolutionary algorithms (EA) operate on a population of potential solutions, applying the principle of survival of the fittest. Genetic algorithms belong to a larger class of EA, which generate solutions to optimization problems using techniques inspired by natural evolution, such as inheritance, mutation, selection, and crossover. This paper proposes a scheduling technique based on genetic algorithm to schedule jobs effectively in a grid. The proposed algorithm is tested with different sizes of preemptive job requests, and analysis of results has shown significant improvement in scheduling performance. Keywords Grid computing Scheduling
Evolutionary algorithms
Genetic algorithm
P. Deepan Babu (&) Department of IT and CT, VLB Janakiammal College of Arts and Science, Coimbatore, Tamil Nadu, India e-mail: [email protected] T. Amudha Department of Computer Applications, Bharathiar University, Coimbatore, Tamil Nadu, India e-mail: [email protected]
G. S. S. Krishnan et al. (eds.), Computational Intelligence, Cyber Security and Computational Models, Advances in Intelligent Systems and Computing 246, DOI: 10.1007/978-81-322-1680-3_42, Springer India 2014
385
386
P. Deepan Babu and T. Amudha
1 Introduction 1.1 Grid Computing Grid computing is an ever-growing area that keeps developing at a constant phase. A grid is a set of resources (such as CPU, memory, disk, applications, and database) distributed over wide area networks and supports large-scale distributed applications [1]. A computational grid is a hardware and software infrastructure that provides dependable, consistent, pervasive, and inexpensive access to highend computational capabilities [2]. Grid architectures are dynamic in nature, any resource can join or leave the grid, and a resource is disparate and connects different networks. Grid computing achieved various breakthroughs in meteorology, physics, medicine, collaborative, or e-science computing [3] and dataintensive computing. A data grid is a major type of grid [4], used in data-intensive applications, where size of data files reaches terabytes or sometimes petabytes. High-energy physics (HEP) and genetic and earth observation are examples of such applications. Data grid is an integrating architecture that connects a group of geographically distributed resources [5]. Computational grid [6] is developed to solve problems that require processing a large quantity of operations. Many research projects require a lot of CPU time, some requires a lot of memory, and some projects need the ability to communicate in real time. Today, sup
Data Loading...
