A Survey on Data Plane Programming with P4: Fundamentals, Advances, and Applied Research

  title={A Survey on Data Plane Programming with P4: Fundamentals, Advances, and Applied Research},
  author={Frederik Hauser and Marco H{\"a}berle and Daniel Merling and Steffen Lindner and Vladimir Gurevich and Florian Zeiger and Reinhard Frank and Michael Menth},

Telemetry and AI-based security P4 applications for optical networks [Invited]

The potentials and challenges of programmable packet-optical nodes encompassing coherent pluggable modules applied in the context of optical metro networks are presented and two innovative applications of P4-based data plane programmability are presented.

Data-driven Method for In-band Network Telemetry Monitoring of Aggregated Traffic

A data-driven in-band telemetry method to monitor the aggregated traffic of the network at the switch level and approximated model parameters provide information for network management for prognostic purposes and congestion avoidance resource planning when integrating new applications.

Attack Detection and Mitigation using Intelligent Data Planes in SDNs

This work uses decision tree and logistic regression based ML models to identify decision boundaries at the controller site and dynamically translates these decision boundaries into range compressed match-action table rules, enabling switches to filter out attack traffic without needing to consult with the SDN controller for each new packet.

Using Programmable P4 Switches to Reduce Communication Costs of Parallel and Distributed Simulations

  • Shie-Yuan WangN. Kuo
  • Computer Science
    GLOBECOM 2022 - 2022 IEEE Global Communications Conference
  • 2022
This work designs and implements a data fusion-based approach inside the packet processing pipeline of the P4 switch that can speed up a PDS by a factor of 2.75 and 1.65 when compared with the unicast and multicast approaches, respectively.

Multi-step Attack Detection and Mitigation Enhancing In-Network Flow Classification

A two-tier flow classification method that integrates initial in-network inference on early subflow metadata with advanced decision support provided by a subsequent machine learning-based analysis step is proposed, allowing for improved accuracies while providing scalable, timely and certain decisions.

A Cache-enabled NDN Forwarding Plane based on Programmable Switches

A cache-enabled NDN forwarding plane using P4 in which massive caching and file transfers are well supported and a dedicated process is designed to take over the cache and communicate with the P4 switch process.

Latency-driven Optimization of Switching Pipeline Design in Network Chips

SPOT is the first automated optimization solution for switching pipeline design in network chips, being utilized in millions of network products of various kinds and saving manual efforts from days to minutes.

In-network angle approximation for supporting adaptive beamforming

This paper proposes a user-assisted in-network method to optimally approximate the angle of arrival by segmenting the cell area into an exponentially binned grid and make use of the advantages offered by programmable data planes and their match-action table (MAT) logic.

A novel programmable software datapath for software-defined networking

It is demonstrated that NIKSS can be used successfully as a P4 software switch, with an acceptable performance gap compared to native eBPF programming, and an extensive performance evaluation is provided, proving that NikSS might be a viable alternative to existing software switches.

Preprint: Open Source Compiling for V1Model RMT Switch: Making Data Center Networking Innovation Accessible

This paper describes the work toward an open-source compiler backend for compiling P4 16 targeted for the V1Model architecture-based programmable switches.



A Fast and Compact Invertible Sketch for Network-Wide Heavy Flow Detection

Fast detection of heavy flows (e.g., heavy hitters and heavy changers) in massive network traffic is challenging due to the stringent requirements of fast packet processing and limited resource

Scaling Hardware Accelerated Network Monitoring to Concurrent and Dynamic Queries With *Flow

This work introduces *Flow, a switch accelerated telemetry system for efficient, concurrent, and dynamic measurement, to carefully partition processing between switch ASICs and application software.

P4air: Increasing Fairness among Competing Congestion Control Algorithms

The solution P4air continuously monitors the properties of all flows that pass through a switch and groups them based on the behavior of the congestion control algorithms used, and applies appropriate measures to suppress the aggressive flows and boost smaller flows.

Implementing AES Encryption on Programmable Switches via Scrambled Lookup Tables

The Scrambled Lookup Table technique is presented for reducing the number of sequential arithmetic operations required for AES encryption, by utilizing the table matching capability available on programmable switches.

Fast Switch-Based Load Balancer Considering Application Server States

A fast switch-based Load Balancer considering Application Server states (LBAS) is proposed to minimize the processing latency at both load balancers and application servers and significantly reduces the response time of requests.

P4ID: P4 Enhanced Intrusion Detection

  • B. LewisM. BroadbentN. Race
  • Computer Science
    2019 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN)
  • 2019
This work presents P4ID, combining a rule parser, stateless and stateful packet processing using P4, and shows that using this technique, it can achieve a significant reduction in traffic being processed by an IDS.

A Case for Data Centre Traffic Management on Software Programmable Ethernet Switches

This paper introduces a novel traffic management algorithm that combines Rate-limited Strict Priority and Deficit round-robin for latency-aware and fair scheduling respectively and presents an implementation on the bmv2 P4 software switch.

Adaptive Weighted Traffic Splitting in Programmable Data Planes

This paper presents a hash-based data structure that achieves adaptive traffic splitting in programmable data planes and shows significant reduction in flow completion time compared to other data-plane load-balancing schemes such as HULA.

Grasp the Root Causes in the Data Plane: Diagnosing Latency Problems with SpiderMon

This work prototypeizes SpiderMon on using the BMv2 model of P4, and the preliminary evaluation shows that SpiderMon is able to quickly find the root cause of performance degradation problems with minimal overhead.

FLAIR: Accelerating Reads with Consistency-Aware Network Routing

The proposed FLAIR leverages the capabilities of the new generation of programmable switches to serve reads from follower replicas without compromising consistency, and implements the processing pipeline using the P4 programming language.