Enhanced Resource Discovery Algorithm for Efficient Grid Computing
Muhyeeddin Alqaraleh
Abstract
In general, parallel applications require lots of computer power and grid computing. Efficient Resource Discovery (RD) algorithms determine grid resource allocation and execution time. To improve the resource distribution and reduce grid node communication costs, this study introduces hierarchical, weighted RD (WRD). A behavioral modeling technique checks the algorithm’s accuracy and efficacy. For complete analysis and simulation, StarUML implements a WRD algorithm behavioral model. The NuSMV model checker evaluates reachability and deadlock-free RD. The WRD algorithm is assessed using key performance metrics. To evaluate resource-finding efficiency, each request’s inspected nodes are counted. The number of re-discovery operations shows the algorithm’s resource flexibility. Algorithms that find free resources with high bandwidth links are also evaluated to optimize grid resource allocation. Resource information tables could improve resource location. Resource information is stored in a table to help the algorithm allocate resources. This research seeks to develop grid computing by solving RD problems. The hierarchical architecture and weighted resource selection of the WRD algorithm improve resource inspection, flexibility, and high-bandwidth RD. Behavioral modeling and verification show the algorithm’s accuracy and grid suitability. WRD and resource information tables boost grid computing RD efficiency and efficacy. This research optimizes grid performance and resource allocation.