Efficient Branch-and-Bound Algorithm for Task Scheduling in Cloud Computing

To fully leverage the potential of cloud computing, it is essential to have efficient scheduling algorithms that can effectively assign tasks to the most appropriate resources for execution. This involves the process of mapping and scheduling tasks onto existing resources in a way that optimizes utilization and minimizes the total time required for completion, known as makespan. Utilization refers to the degree to which the network resources of the cloud are utilized. However, due to the dynamic and heterogeneous nature of cloud resources, task scheduling is a challenging NP-complete problem that requires the use of metaheuristics to identify the optimal scheduling approach. However, the use of an exact method in this context is rarely explored. The main objective of this paper is to optimize the task scheduling to minimize the makespan by using a new bounding operator that improves the scalability of the standard B&B algorithm. The proposed approach has been compared with the cuckoo search methaheuristic using the cloudsim simulator.