Showing papers on "Degree of parallelism published in 2022"
01 Jan 2022
TL;DR: In this paper, a hybrid load balancing algorithm is proposed to handle the client requests based on the priority and availability of the cluster nodes, which are part of the virtual system, and maintains two tables with a list of node id, priority, and availability so that load can be optimally distributed.
Abstract: When the number of processes requiring CPU is higher than the number of cores in a system, achieving parallel systems show many challenges. However, different approaches are known to work well for achieving an acceptable degree of parallelism. Load balancing operations are one of the main concerns in virtual systems and, specifically, in Cloud Computing. An optimal load balancing strategy can increase the multiprogramming rate, significantly impacting the parallelism degree of a system. There are many algorithms proposed and designed to carry out client requests at distributed servers in the cloud. All of them target maximizing the performance of the overall system. In this manuscript, various existing load balancing algorithms are discussed and compared with a new hybrid algorithm. This hybrid algorithm has been developed to handle the client requests based on the priority and availability of the cluster nodes, which are part of the virtual system. This algorithm maintains two tables with a list of node id, priority, and availability so that the load can be optimally distributed. That enables selecting an appropriate node to achieve efficient load balancing and therefore s a higher degree of overall parallelism.