Cost-effective clonal selection and AIS-based load balancing in cloud computing environment.

Cloud computing faces challenges in increasing efficiency through task scheduling and load balancing. Various load-balancing methods, including the Throttled policy, help evenly distribute user requests among virtual machines. The article compares the cost of the clonal selection algorithm with the throttled load-balancing policy and the clonal selection algorithm with the distributed service broker policy. It also introduces two new algorithms based on the artificial immune system, aiming to efficiently respond to user requests at the lowest cost. Additionally, the article presents two new algorithms resulting from combining an evolutionary algorithm with artificial immune system mechanisms. These algorithms have been evaluated and compared using the Cloud Analyst simulator, demonstrating their superior performance in terms of response time, data center processing time, data transfer cost, and overall cost. The results indicate that ACODTH algorithm has an average response time of 0.35 milliseconds shorter than ACO. Additionally, the ACONSA algorithm shows a 0.44-millisecond reduction in average data center processing time compared to ACONSA. When considering the overall cost, the AIS_DSBP algorithm is projected to be 3.13$ cheaper than AIS_TLP. Similarly, the CSA_DSBP algorithm is expected to have a total data transfer cost of 3.14$ lower than CSA. [ABSTRACT FROM AUTHOR]