A Novel Resource Scheduling Algorithm for Computational Grid

Abstract

Grid is an emerging technology for enabling resource sharing and coordinated problem solving in dynamic multiinstitutional heterogeneous virtual organizations. However, the management of resources and computational tasks is a critical and complex undertaking as these resources and tasks are geographically distributed. A suitable and efficient scheduling algorithm is needed to schedule user tasks to heterogeneous resources distributed in the grid. We propose a genetic algorithm to solve the scheduling problems in computational grid. The proposed algorithm uses the optimal searching technique of genetic approach and takes different computing capabilities of nodes into consideration. The simulation result shows that proposed algorithm yields better performance when compared with other traditional heuristic approaches.

Modified Hierarchical Method for Task Scheduling in Grid Systems

“…Task scheduling in grid systems is an NP-complete problem [5,6], and the solution has different approximate methods and algorithms, such as heuristic algorithms [7], genetic algorithms [8,9,10], algorithms based on stochastic Petri networks [11], ant colony algorithms [12], fuzzy optimization [13], tabu search [14], gravitational emulation local search [15], learning automata [16] and combinations thereof [17,18,19,20]. In the general case, task scheduling is a multi-objective problem in grid systems.…”

Application of #Chain Equations to Resource Scheduling

“…They include the Earliest Start (ES), Latest Start (LS) and Resource Leveling (RL) methods, the Tabular Graphical Resource (TGR) method, and applications of Genetic Algorithms (GA) for resource scheduling. GA has been employed to solve scheduling problems in computational grid [7,8,9]. The RL method requires that two resource profiles, the ES and the LS resource profiles should be present before the resource leveling can be performed.…”