CNS Core: Small: Economic Optimization of Serverless and VM Cloud Services

点击翻译按钮获取中文摘要

Cloud computing platforms are responsible for the ever-increasing computation in the United States and the world. Its uses extend throughout private industry, government, and academia. These cloud platforms are offered to users via a plethora of computing paradigms and pricing models. While traditionally, users would pay for the reservation of a computational resource, a recently evolving paradigm, known as Function-as-a-Service (FaaS) or as Serverless Computing (SC), has emerged, where users pay instead only for the time their compute jobs are executing. This has had significant implications on how these systems are priced, which in turn affects how they are utilized. This research will use a combination of mathematical modeling, simulation, and prototyping to formally understand the impact that serverless computing has on cloud computing use, both from an economic and engineering perspective. The research applies a game theoretic formulation in which players are rational agents who are trying to maximize their own individual welfare. The results will show how cloud computing markets are expected to evolve to accommodate the serverless computing option. More generally, the results of the research will provide a well-grounded foundation for improved understanding as to how pricing strategies for these differing paradigms impact the use and profitability of both Cloud computing providers and consumers.The research will expand the nation's understanding of the emerging computational systems that are widely utilized, and will continue to have increasing demand. From both an economic and engineering perspective, the research will enable a more effective use of such systems, which in turn will increase the efficiency and effectiveness of the nation's computational infrastructure. The theoretical style of the research is ideal for effective conveyance of the fundamental issues of the topic, making it an important and topical addition to course curriculum. The PIs are committed to broadening participation in this exciting field of research. In particular, they have and will continue to reach out to female students and under-represented minority groups. The project will actively encourage undergraduate research.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.