Developing High Performance Networked Applications using P4-DPDK

Overview

Network packet processing faces significant performance challenges due to kernel overheads. These issues have become more pronounced with the rapid growth of network traffic. To address these challenges, the Data Plane Development Kit (DPDK) was developed. DPDK bypasses the kernel and operates directly in user space, offering significant improvements in performance and latency for packet processing tasks. However, DPDK's steep learning curve presents a barrier to entry for developers and network administrators. In recent years, P4 has emerged as a language specifically designed for expressing packet processing data paths. Building on this development, P4-DPDK has been introduced as a new technology that bridges P4 and DPDK. It allows developers to create P4 code which is then translated into a DPDK pipeline, combining the expressiveness of P4 with the performance benefits of DPDK. This workshop aims to provide researchers, students, developers, and practitioners with an introduction to P4-DPDK, followed by a hands-on lab. The workshop will cover the fundamentals of P4, including P4 building blocks, parser implementation, and match-action tables.

Audience

The target audience for this tutorial includes Information Technology (IT) professionals and practitioners, researchers, and application developers. The tutorial can also be useful for instructors interested in teaching P4 in the classrooms.

Tutorial Goals

The workshop aims to equip participants with both theoretical knowledge and practical skills related to packet processing in high-speed networks. By the end of this workshop, attendees will:

• Implement DPDK pipelines using the P4 language.
• Describe the elements of the Portable NIC Architecture (PNA), define and parse protocol headers and header fields in P4, define match-action tables and populate them at runtime.
• Develop DPDK-based defense solutions.
• Accelerate packet processing using the Receive Side Scaling (RSS), which distributes packets across multiple CPU cores.

Pre-requisites

Connectivity to the Internet and a browser to access the online virtual platform. Attendees will be provided with an account to access USC’s NETLAB system: https://netlab.cec.sc.edu/

Agenda

Tuesday, March 11