Search for: All records

Software-defined networking (SDN) in conjunction with programmable switches revolutionizes network management, yet crafting optimal switch configurations remains complex. Traditional P4 optimizations rely on data plane level tuning. In this paper, we argue an essential piece for such optimizations is the control plane itself. We present P4CGO, a P4 compilation framework which focuses on realizing specifications based on control policies. P4CGO leverages user-defined objective functions and control plane policies to guide P4 program optimization through table merging and splitting. We have prototyped P4CGO and applied it solving real-world policy optimization problems. 

Online traffic classification enables critical applications such as network intrusion detection and prevention, providing Quality-of-Service, and real-time IoT analytics. However, with increasing network speeds, it has become extremely challenging to analyze and classify traffic online. In this paper, we present Leo, a system for online traffic classification at multi-terabit line rates. At its core, Leo implements an online machine learning (ML) model for traffic classification, namely the decision tree, in the network switch's data plane. Leo's design is fast (can classify packets at switch's line rate), scalable (can automatically select a resource-efficient design for the class of decision tree models a user wants to support), and runtime programmable (the model can be updated on-the-fly without switch downtime), while achieving high model accuracy. We implement Leo on top of Intel Tofino switches. Our evaluations show that Leo is able to classify traffic at line rate with nominal latency overhead, can scale to model sizes more than twice as large as state-of-the-art data plane ML classification systems, while achieving classification accuracy on-par with an offline traffic classifier.