Unlocking the Power of Inline Floating-Point Operations on Programmable Switches

Yifan Yuan, Omar Alama, Jiawei Fei, Jacob Nelson, Dan R.K. Ports, Amedeo Sapio, Marco Canini, Nam Sung Kim

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

19 Scopus citations

Abstract

The advent of switches with programmable dataplanes has enabled the rapid development of new network functionality, as well as providing a platform for acceleration of a broad range of application-level functionality. However, existing switch hardware was not designed with application acceleration in mind, and thus applications requiring operations or datatypes not used in traditional network protocols must resort to expensive workarounds. Applications involving floating point data, including distributed training for machine learning and distributed query processing, are key examples. In this paper, we propose FPISA, a floating point representation designed to work efficiently in programmable switches. We first implement FPISA on an Intel Tofino switch, but find that it has limitations that impact throughput and accuracy. We then propose hardware changes to address these limitations based on the open-source Banzai switch architecture, and synthesize them in a 15-nm standard-cell library to demonstrate their feasibility. Finally, we use FPISA to implement accelerators for training for machine learning as an example application, and evaluate its performance on a switch implementing our changes using emulation. We find that FPISA allows distributed training to use one to three fewer CPU cores and provide up to 85.9% better throughput than SwitchML in a CPU-constrained environment.

Original languageEnglish (US)
Title of host publicationProceedings of the 19th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2022
PublisherUSENIX Association
Pages683-700
Number of pages18
ISBN (Electronic)9781939133274
StatePublished - 2022
Event19th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2022 - Renton, United States
Duration: Apr 4 2022Apr 6 2022

Publication series

NameProceedings of the 19th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2022

Conference

Conference19th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2022
Country/TerritoryUnited States
CityRenton
Period04/4/2204/6/22

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Control and Systems Engineering

Fingerprint

Dive into the research topics of 'Unlocking the Power of Inline Floating-Point Operations on Programmable Switches'. Together they form a unique fingerprint.

Cite this