Smart Parallel Algorithm for Optimized Load Balancing in Cloud Computing

On demand scalable distributed resources over a distributed network, cloud computing completely changed the face of modern computing. Despite the workloads in the cloud being dynamic, resource heterogeneous and requiring real time task allocation, load balancing is still a critical challenge. Most traditional load balancing approaches face high computational overhead, not fully utilizing resource and are not easily adaptable to dynamic cloud environments. To improve efficiency in cloud computing, this paper introduces Smart Parallel Algorithm based Load Balancing (SPALB) framework. The proposed approach utilizes hybrid parallel computing algorithm by utilizing the swarm intelligence coupled with meta-heuristic optimization and adaptive scheduling schemes to achieve the optimal load distribution. In contrast to traditional methods, SPALB performs workload balancing dynamically using multi objective optimization, achieves task execution deadlines and minimizes energy consumption with error detection and correction. The framework is also implemented and evaluated on cloud simulation platforms and real clouds including AWS and Openstack. Throughput, response time, resource utilization as well as scalability results of the experimental results show that SPALB outperforms all existing load balancing techniques. The proposed approach further improves fault tolerance and security in multi tenant cloud environments and can resist against the cyber threats. Although capable of circumventing the limitations of traditional cloud load balancing, smart parallel algorithms carry limitations and the study shows the potential of smart parallel algorithms for future research on AI driven adaptive load balancing algorithms.