An Improved Crow Search Load Balancing Algorithm Using ThresholdBased Strategy for Virtual Machine Allocation
Mathias Pama Garkuwa, Rabi Mustapha, Khalid Haruna
Keywords: Cloud computing, Crow Search, Virtual Machine, Load balancing Algorithms, Threshold-Based Strategy
Load balancing is the allocation (distribution) of workload among a set of computational elements (CEs). In large-scale distributed computing systems, CEs are physically or virtually distant from each other. Communication among CEs is associated with delays that can significantly alter the expected performance of load-balancing policies. This is a prominent problem in computing systems for which the individual units are connected by means of a shared communication medium such as the Internet, ad-hoc networks, wireless Local Area Networks. Aim: This study is aimed at providing an improved Crow search load balancing algorithm (ICSLBA) to enhance performance, simulate and evaluate the proposed algorithm using some performance metrics. Method: the existing Crow Search Algorithm is improved by introducing a Threshold-based strategy and simulated using a cloud simulator with an aim to enhance resource allocation in a cloud computing environment. Results: The results achieved through experimentation have shown that; the proposed ICSLBA has out-performed the CSLBA in terms of improving Average Makespan Time (AMT), Average Waiting Time (AWT), and Average Data Center Processing Time (ADCPT). Hence, the proposed ICSLBA is the optimal load-balancing strategy.