Litcius/Paper detail

FLASH: FPGA-Accelerated Smart Switches with GCN Case Study

Pouya Haghi, William Krska, Cheng Tan, Tong Geng, Po Hao Chen, Connor Greenwood, Anqi Guo, Thomas Hines, Chunshu Wu, Ang Li, Anthony Skjellum, Martin Herbordt

202311 citationsDOI

Abstract

Some communication switches, e.g., the Mellanox SHArP and those in the IBM BlueGene clusters, are augmented to process packets at the application level with fixed-function collectives. This approach, however, lacks flexibility, which limits their applicability in diverse and dynamic workloads. Recently, a new type of programmable packet processor, which uses high-level languages, e.g., P4, has emerged as a possible candidate. P4-based switches, however, fall short in certain applications, including machine learning, where capabilities not currently supported by P4 are needed. These include more complex calculation, such as sparse computation and fused multiply-accumulate, data-intensive floating point operations, data reuse, and significant memory. The problem addressed here is that such a switch augmentation needs to support: a large amount of state, significant flexible compute capability, and ease of programming, all while maintaining full functionality, including ensuring high throughput, and demonstrating utility.

Topics & Concepts

Computer scienceField-programmable gate arrayFlexibility (engineering)ThroughputReuseNetwork packetEmbedded systemIBMState (computer science)Process (computing)Remote direct memory accessComputer architectureComputer hardwareOperating systemComputer networkProgramming languageWirelessEcologyStatisticsNanotechnologyBiologyMathematicsMaterials scienceSoftware-Defined Networks and 5GInterconnection Networks and SystemsRadiation Effects in Electronics
FLASH: FPGA-Accelerated Smart Switches with GCN Case Study | Litcius