Performance Evaluation of the Virtualization Environment of a Microservices-Based Payroll System
Klayton Rodrigues de Castro, Lucas M. C. e Martins, Polyane Wercelens, Rafael Padilha, Italo Gervasio, Flávio Deus, William Ferreira Giozza, Rafael T. de Sousa · 2020
For better usage of idle resources in a symmetric multiprocessing environment, cloud computing providers often exploit the boundaries of parallelism by imposing high CPU subscription rates over their virtualization systems. Moreover, unsuitable resource allocation can significantly impair the performance during intensive workloads and increase infrastructure expenditures unnecessarily. It becomes an increasing challenge when dealing with microservices architecture and container encapsulated applications, for these are approaches that add more intricacy layers to the workload scheduling over the hardware. This paper presents a case study on a public agency private cloud, which aims to evaluate the performance of hypervisors and virtual machines, that are serving a Kubernetes container orchestration cluster and running a microservices-based Payroll System during CPU-bound behavior tasks. We also propose a model for sizing the virtual environment according to the best choices for the hardware characteristics and the workload needs.