Audition: a DevOps-oriented service optimization and testing framework for cloud environments

Abstract

This paper demonstrates an approach to automated testing and quality assurance in cloud environments, which also takes deployment cost into consideration. With a distributed service architecture and some given performance goals, the end result will be a suggestion of the optimal resource type and filesystem with the lowest price point for each function of the architecture. Our solution is modeled after the auditioning process in the theater industry, which provides a process that fits well into our context and is easy to understand and follow. The resulting tool, Audition, is a working implementation of our model and is extendable in several ways, allowing for integration with local technologies.