Taurus: a data plane architecture for per-packet ML
Authors:
Top Cited Papers
Open Access
- 22 February 2022
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1099-1114
Abstract
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 data-driven 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 line-rate 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.Keywords
All Related Versions
Funding Information
- Defense Advanced Research Projects Agency (FA8650-18-2-7865)
- NSF (National Science Foundation) (1937301)
This publication has 82 references indexed in Scilit:
- Support Vector Machines for TCP traffic classificationComputer Networks, 2009
- McPAD: A multiple classifier system for accurate payload-based anomaly detectionComputer Networks, 2009
- CUBICACM SIGOPS Operating Systems Review, 2008
- NN-RED: an AQM mechanism based on neural networksElectronics Letters, 2007
- An Adaptive Neuron AQM for a Stable Internet2007
- Traffic classification on the flyACM SIGCOMM Computer Communication Review, 2006
- An improved data stream summary: the count-min sketch and its applicationsJournal of Algorithms, 2005
- Long Short-Term MemoryNeural Computation, 1997
- Random early detection gateways for congestion avoidanceIEEE/ACM Transactions on Networking, 1993
- Approximation capabilities of multilayer feedforward networksNeural Networks, 1991