Hybrid Grey Wolf and Particle Swarm Optimized (HGW-PSO) Scheduling in Containerized Cloud Computing Environment

The concept of containerization consists of encapsulating applications with their libraries with the help of operating system level virtualization, ensuring they are separated from other processes on the same machine. Containers gained popularity in several data centers due to their lightweight nature and simplified deployment. Application containerization technology is a fundamental technology that enables the deployment of numerous containers on a single physical node. A single physical node has the capability to offer several services to users. The adoption of microservice architecture is on the rise as it offers several benefits over monolithic designs, such as simplified deployment, enhanced versatility, and improved accessibility. The rise of containerization has significantly increased the significance of microservices, enhancing their scalability and portability. Many researchers have developed theories and solutions for the microservice container scheduling problem, yet limits remain. The algorithm's tendency to become stuck in local optimization and its slow optimal solution search are examples. In this work we present a proposal for grey wolf and particle swarm optimized (GW-PSO) scheduling in a Containerized Cloud Computing Environment. During the first phase of the search, the GW's global search can direct the PSO away from local optima, preventing early convergence. In the last phases of the search, the PSO can improve GW search along the Pareto optimum frontier. The research findings demonstrate, in comparison to the other 3 methods, the algorithm achieves a 19.06% reduction in network transmission cost, a 15.57% improvement in load balancing performance, and a 6.67% enhancement in service stability.