Serverless computing is gaining traction as an attractive model for the deployment of a multitude of workloads in the cloud. Designing and building effective resource management solutions for any computing environment requires extensive long term testing, experimentation and analysis of the achieved performance metrics. Utilizing real test beds and serverless platforms for such experimentation work is often times not possible due to resource, time and cost constraints. Thus, employing simulators to model these environments is key to overcoming the challenge of examining the viability of such novel ideas for resource management. Existing simulation software developed for serverless environments lack generalizibility in terms of their architecture as well as the various aspects of resource management, where most are purely focused on modeling function performance under a specific platform architecture. In contrast, we have developed a serverless simulation model with induced flexibility in its architecture as well as the key resource management aspects of function scheduling and scaling. Further, we incorporate techniques for easily deriving monitoring metrics required for evaluating any implemented solutions by users. Our work is presented as CloudSimSC, a modular extension to CloudSim which is a simulator tool extensively used for modeling cloud environments by the research community. We discuss the implemented features in our simulation tool using multiple use cases.

翻译：Serverless计算正逐渐成为一种吸引人的云计算工作负载部署模型。针对任何计算环境设计并构建有效的资源管理解决方案，都需要对性能指标进行长期测试、实验与分析。由于资源、时间和成本的限制，利用真实测试平台和Serverless平台进行此类实验往往不可行。因此，使用仿真器对这些环境进行建模，是克服检验资源管理创新方案可行性这一挑战的关键。现有面向Serverless环境的仿真软件在架构及资源管理的多个方面缺乏通用性，大多数软件仅专注于特定平台架构下的函数性能建模。相比之下，我们开发了一种在架构以及函数调度与扩缩容等关键资源管理方面具有灵活性的Serverless仿真模型。此外，我们引入了相关技术，以便用户轻松推导出评估任何已实施方案所需的监控指标。我们的工作以CloudSimSC的形式呈现，它是CloudSim的一个模块化扩展，CloudSim是学术界广泛用于云环境建模的仿真工具。我们通过多个用例讨论了仿真工具中实现的功能。