Serverless computing becomes the new trending paradigm in cloud computing, allowing developers to focus on the core application logic and rapidly prototype applications. Due to the great prospects of serverless computing, in recent years, most major cloud vendors have rolled out their commodity serverless computing platforms. However, the characteristics of these platforms have not been systematically studied. To fill this knowledge gap, this paper presents a comprehensive study on characterizing mainstream commodity serverless computing platforms (i.e., AWS Lambda, Azure Functions, Google Cloud Functions, and Alibaba Cloud Function Compute). First, we qualitatively analyze these platforms from development, deployment, and runtime aspects to form the taxonomy of characteristics. Additionally, we quantitatively evaluate the actual performance of different serverless computing platforms through our designed benchmark platform. Our benchmark platform has two types of benchmarks, i.e., microbenchmarks and macrobenchmarks. Based on the results of qualitative and quantitative analyses, we derive a series of findings and provide insightful implications for both developers and cloud vendors.

翻译：无服务器计算成为云计算中的新趋势模式,使开发商能够关注核心应用逻辑和快速原型应用。由于无服务器计算的巨大前景,近年来,大多数大云供应商都推出了无服务器计算平台。然而,这些平台的特征尚未系统地研究。为填补这一知识空白,本文件全面研究了无服务器计算平台(如AWS Lambda、Azure函数、Google云函数和Alibaba Cloud函数计算)的特点。首先,我们从质量角度分析了这些平台的发展、部署和运行时间方面,以形成特征分类。此外,我们还通过我们设计的基准平台对不同无服务器计算平台的实际性能进行了定量评估。我们的基准平台有两种基准,即微型基准和宏观基准。根据定性和定量分析的结果,我们得出了一系列结论,并为开发商和云供应商提供了深刻的影响。