The documents distributed by this server have been provided by the contributing authors as a means to ensure timely dissemination of scholarly and technical work on a noncommercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.
Publications of SPCL
|M. Copik, R. Böhringer, A. Calotoiu, T. Hoefler:|
|FMI: Fast and Cheap Message Passing for Serverless Functions|
AbstractServerless functions provide elastic scaling with a fine-grained billing model. However, for distributed jobs that benefit from large-scale and dynamic parallelism, the lack of fast and cheap communication is a major limitation of FaaS computing. Individual functions do not communicate directly and group operations do not exist, and users resort to manual implementations of storage-based communication. We overcome this limitation and present the FaaS Message Interface (FMI). FMI is an easy-to-use, high-performance framework for general-purpose point-to-point and collective communication. We support different communication channels and offer a model-driven channel selection according to performance and cost expectations. We model the interface after MPI and show message passing can be integrated into serverless applications with minor changes. In our experiments, FMI can speed up communication for a distributed machine learning job by up to 162x while at the same time reducing cost by up to 941 times..