Elsevier

Future Generation Computer Systems

Volume 124, November 2021, Pages 268-284
Future Generation Computer Systems

Benchmarking parallelism in FaaS platforms

https://doi.org/10.1016/j.future.2021.06.005Get rights and content
Under a Creative Commons license
open access

Highlights

  • FaaS is not inherently good for parallel computations, depending on implementation.

  • Parallel performance changes noticeably between Cloud FaaS platforms.

  • AWS and IBM support parallelism; Azure and in less degree Google present problems.

  • Parallelism improves with light virtualization and push-based architectures.

Abstract

Serverless computing has seen a myriad of work exploring its potential. Some systems tackle Function-as-a-Service (FaaS) properties on automatic elasticity and scale to run highly-parallel computing jobs. However, they focus on specific platforms and convey that their ideas can be extrapolated to any FaaS runtime.

An important question arises: do all FaaS platforms fit parallel computations? In this paper, we argue that not all of them provide the necessary means to host highly-parallel applications. To validate our hypothesis, we create a comparative framework and categorize the architectures of four cloud FaaS offerings, emphasizing parallel performance. We attest and extend this description with an empirical experiment that consists in plotting in deep detail the evolution of a parallel computing job on each service.

The analysis of our results evinces that FaaS is not inherently good for parallel computations and architectural differences across platforms are decisive to categorize their performance. A key insight is the importance of virtualization technologies and the scheduling approach of FaaS platforms. Parallelism improves with lighter virtualization and proactive scheduling due to finer resource allocation and faster elasticity. This causes some platforms like AWS and IBM to perform well for highly-parallel computations, while others such as Azure present difficulties to achieve the required parallelism degree. Consequently, the information in this paper becomes of special interest to help users choose the most adequate infrastructure for their parallel applications.

Keywords

Serverless
FaaS
Parallelism
Benchmark

Cited by (0)

Daniel Barcelona-Pons is a Ph.D. student at the Computer Engineering and Mathematics Department of the Universitat Rovira i Virgili (Spain). His research interests include cloud computing, distributed systems, and middleware. His recent work explores serverless computing with publications on prestigious conferences such as ACM Middleware.

Pedro García-López is professor of the Computer Engineering and Mathematics Department at the University Rovira i Virgili (Spain). He leads the ”Cloud and Distributed Systems Lab” research group and has coordinated large European research projects: FP7 CloudSpaces, H2020 IOStack and CloudButton.

He worked as visiting scientist at IBM Watson Research. His research topics are distributed systems, cloud computing, data analytics, software architectures and middleware. He has published more than 100 papers on journals and prestigious conferences and participated in scientific committees of different conferences like Middleware and CCGRID, among others. He is currently co-organizing the International Workshop on Serverless Computing.