SPAC: Automating FPGA-based Network Switches with Protocol Adaptive Customization

TL;DR

SPAC automates the design of FPGA-based network switches tailored to specific protocols and traffic patterns, optimizing performance and resource usage for diverse applications.

Contribution

It introduces a unified workflow with a DSL, adaptive components, and a trace-aware DSE engine for automated, protocol-aware FPGA switch customization.

Findings

SPAC reduces LUT and BRAM usage by over 50%.

Latency improvements of up to 38.4% are achieved.

Designs are effective across diverse real-world scenarios.

Abstract

With network requirements diverging across emerging applications, latency-critical services demand minimal logic delay, while hyperscale training and collectives require sustained line-rate throughput for synchronized bulk transfers. This divergence creates an urgent need for custom network switches tailored to specialized protocols and application-specific traffic patterns. This paper presents SPAC (Switch and Protocol Adaptive Customization), a novel approach that automates the generation of FPGA-based network switches co-optimized for custom protocols and application-specific traffic patterns. SPAC introduces a unified workflow with a domain-specific language (DSL) for protocol-architecture co-design, a library of modular HLS-based adaptive switch components, and a trace-aware Design Space Exploration (DSE) engine. By providing a multi-fidelity simulation stack, SPAC enables rapid identification of Pareto-optimal designs prior to deployment. We demonstrate the efficacy of the domain-specific adaptation of SPAC across a spectrum of real-world scenarios, spanning from latency-sensitive sensor and HFT networks to hyperscale datacenter fabrics. Experimental results show that by tailoring the micro-architecture and protocol to the specific workload, SPAC-generated designs reduce LUT and BRAM usage by 55% and 53%, respectively. Compared to fixed-architecture counterparts, SPAC delivers latency reductions ranging from 7.8% to 38.4% across various tasks while maintaining adequate resource consumption and packet drop rate.