Energy-Aware Microservice-Based SaaS Deployment in a Cloud Data Center Using Hybrid Particle Swarm Optimization

Abstract

The deployment of software as a service (SaaS) using a microservice architecture offers several benefits, including scalability, flexibility, and ease of maintenance. One of the most important advantages of the new microservice-based SaaS deployment is that the increase in energy consumption incurred by the deployment of a new microservice-based SaaS can be considered. With the aim of reducing the increase in energy consumption, this paper proposes a new method, namely Hybrid Particle Swarm Optimization (HPSO), to solve the microservice-based SaaS deployment problem. The HPSO incorporates adaptive inertia weight, cognitive, and social parameters to balance the trade-off between exploration and exploitation during the optimization process. Furthermore, the HPSO incorporates a local optimizer to improve the best global solution within the swarm, with a specific emphasis on improving energy efficiency. To evaluate the performance of the HPSO, we have implemented it and compared it with a GA method by experiment. The experimental results have shown that the HPSO can further reduce the increase in energy consumption by 3.68% compared to GA.

Keywords

• Cloud computing
• data center
• deployment
• energy consumption
• hybrid particle swarm optimization
• microservice