Containerized Deployment of Microservices in Cloud Computing

Abstract

Containerization results in the proliferation of several benefits to cloud computing due to its speed, scalability, and security features. Additionally, the growing popularity of containerizing sophisticated applications (such as Netflix, Spotify, and the Times of India) and deploying them over the cloud has aided businesses in the development, testing, management, deployment, and security of their software and applications. This increase in popularity, embarked demand for research in the areas of efficiency and optimization of metrics such as resources (CPU, memory, storage, and network), power consumption, and cost, to make the deployment more feasible and efficient in developing and scaling services of the applications. This work proposed heuristics and PSO-based approach to optimize microservices-to-container placement and container-to-server placement, respectively. The proposed methods efficiently utilize resources and power, thus providing cost efficiency to the deployment and auto-scaling of the microservices of an application. More specifically, dynamic bin packing and Particle-swarm optimization techniques are adopted as optimization algorithms for minimizing resource wastage and power consumption. The proposed algorithms are then compared with state-of-the-art and the results verify the effectiveness of the proposed work.