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

The advent of switches with programmable dataplanes has enabled the rapid development of newnetwork 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.

We would like to thank our shepherd, Ellen Zegura, and the anonymous reviewers for their helpful feedback. We also thank Zhe Chen, Muhammad Tirmazi, and Minlan Yu for their technical support and discussion. This research is partially supported by National Science Foundation (No. CNS-1705047), by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research (OSR) under Award No. OSR-CRG2020-4382, and by a gift in kind from Huawei. For computer time, this research used the resources of the Supercomputing Laboratory at KAUST. This researchwas partially done when the first authorwas at Microsoft Research. The work of Jiawei Fei at KAUST is supported by a sponsorship from China Scholarship Council (CSC).


Conference/Event Name
19th USENIX Symposium on Networked Systems Design and Implementation (NSDI)


Additional Links

Permanent link to this record

Version History

Now showing 1 - 2 of 2
2022-12-01 08:23:50
Published as conference paper
2022-05-17 13:27:58
* Selected version