Coarse-Grained Reconfigurable Architectures

Prabhakar, Raghu; Zhang, Yaqi; Olukotun, Kunle

doi:10.1007/978-3-030-18338-7_14

Cited by 1 publication

(1 citation statement)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Target-Independent Optimizations. MapReduce is general enough to support target-independent optimizations: optimizations that consider available execution resources (parallelization factors, bandwidth, and more) without considering hardware-specific design details [128,176]. Parallelizing MapReduce programs unrolls loops in space: if sufficient hardware resources are available, a model can execute one iteration per cycle.…”

Section: Taurus Implementationmentioning

confidence: 99%

Taurus: a data plane architecture for per-packet ML

Swamy

Rucker

Shahbaz

et al. 2022

Proceedings of the 27th ACM International Conference on Architectural Support for Programming Languages and Operating Systems

Self Cite

View full text Add to dashboard Cite

Emerging applications-cloud computing, the internet of things, and augmented/virtual reality-demand responsive, secure, and scalable datacenter networks. These networks currently implement simple, per-packet, data-plane heuristics (e.g., ECMP and sketches) under a slow, millisecond-latency control plane that runs datadriven performance and security policies. However, to meet applications' service-level objectives (SLOs) in a modern data center, networks must bridge the gap between line-rate, per-packet execution and complex decision making.In this work, we present the design and implementation of Taurus, a data plane for line-rate inference. Taurus adds custom hardware based on a flexible, parallel-patterns (MapReduce) abstraction to programmable network devices, such as switches and NICs; this new hardware uses pipelined SIMD parallelism to enable per-packet MapReduce operations (e.g., inference). Our evaluation of a Taurus switch ASIC-supporting several real-world models-shows that Taurus operates orders of magnitude faster than a server-based control plane while increasing area by 3.8% and latency for linerate ML models by up to 221 ns. Furthermore, our Taurus FPGA prototype achieves full model accuracy and detects two orders of magnitude more events than a state-of-the-art control-plane anomaly-detection system.

show abstract

Section: Taurus Implementationmentioning

confidence: 99%