SHF: SMALL: Collaborative Research: Cloud Mentoring: Guiding Cloud Users for Cost Performance through Testing and Recommendation

云计算正在迅速发展，企业、机构和个人将其工作负载转移到云。云用户受益于低成本所有权，即用即付的定价模式，他们只为采购的资源使用付费，以及动态扩展资源使用的能力。然而，在云中运行的应用程序通常会遇到不可预测的性能，这使得用户选择满足其成本和性能要求的资源配置极具挑战性。由于用户对云计算计算机没有物理控制权，并且被迫根据复杂的云性能报告做出决策，因此这个问题变得更加复杂。本研究解决了支持用户在将应用程序移植到各种云服务时实现其成本性能要求的需求。特别是，研究体现在一个设想的测试和推荐系统，确定适当的资源管理政策，满足性能和成本要求。通过采用基于软件测试的方法，该研究提供了仅使用用户可访问的信息来满足用户需求的解决方案，解决了依赖于性能可预测性的静态分析技术的局限性。这样的测试和推荐系统使非专家能够以经济有效的方式将其应用程序移植到各种云。该项目开发的新框架、测试和推荐方法、数据集和实验基础设施将开源发布，以促进软件工程和云计算领域的知识和理解。PI将继续让代表性不足的群体的学生参与，并继续参与为学生举办的辅导讲习班。

Cloud computing is growing rapidly, with businesses, institutions and individuals moving their workloads to clouds. Cloud users benefit from low cost ownership, a pay-as-you-go pricing model where they only pay for the procured resource usage, and the ability to dynamically scale the resource usage up and down. However, the applications running in clouds usually experience unpredictable performance, which makes it extremely challenging for the users to choose resource configurations that meet their cost and performance requirements. This problem is further complicated as users do not have physical control over cloud computers and are forced to make their decisions based on convoluted cloud performance reports.This research addresses the need to support users in achieving their cost-performance requirements as they port their applications to various cloud services. In particular, the research is embodied in an envisioned testing and recommendation system that determines proper resource management policies that meet performance and cost requirements. By taking a software-testing-based approach, the research provides solutions using only user-accessible information to satisfy user requirements, addresses the limits of static analysis techniques that rely on performance predictability.Such a testing and recommendation system enables non-experts to port their applications to various clouds in a cost effective way. The novel framework, testing and recommendation approaches, data sets, and experimental infrastructure developed within the project will be released open source to advance knowledge and understanding within software engineering and cloud computing. The PIs will continue to involve students of underrepresented groups and continue their involvement in mentoring workshops for students.