As developers, how to predict and optimize the performance of your serverless functions? Our paper “Enhancing Performance Modeling of Serverless Functions via Static Analysis” provides you the possibility to build your own analytical performance model for serverless workflows. The proposed method focuses on helping developers to extract model topology by working on their source code, and devises an instrumentation strategy for code-level profiling to enhance the model accuracy. Accurate prediction of response time can be achieved with error rate below 7.3%!
Authors:
Runan Wang is currently pursuing the Ph.D. degree in Department of Computing, Imperial College London. Her research focuses on performance models, program analysis, and serverless computing.
Giuliano Casale is a Reader at Imperial College London. He teaches and does research in performance engineering and cloud computing and has published more than 150 refereed papers.
Antonio Filieri is an associate professor at Imperial College London, and a senior applied scientist with Amazon AWS. His research focuses on application of mathematical method for software engineering.
