DPF Book Template - RDG for DPF with OVN-Kubernetes and HBN Services Demo

Introduction

The NVIDIA BlueField-3 data processing unit (DPU) is a 400 Gb/s infrastructure compute platform designed for line-rate processing of software-defined networking, storage, and cybersecurity. BlueField-3 combines powerful computing, high-speed networking, and extensive programmability to deliver hardware-accelerated, software-defined solutions for demanding workloads.

NVIDIA DOCA unlocks the full potential of the NVIDIA BlueField platform, enabling rapid development of applications and services that offload, accelerate, and isolate data center workloads.

Host-based Networking (HBN) is a DOCA service that allows network architects to design networks based on layer-3 (L3) protocols. HBN enables routing to run on the server side by using BlueField as a BGP router. The HBN solution encapsulates a set of network functions inside a container, which is deployed as a service pod on BlueField's Arm cores.

OVN-Kubernetes is a Kubernetes CNI network plugin that provides robust networking for Kubernetes clusters. Built on Open Virtual Network (OVN) and Open vSwitch (OVS), it supports hardware acceleration to offload OVS packet processing to NIC/DPU hardware. With OVS-DOCA, an extension of traditional OVS-DPDK and OVS-Kernel, accelerated OVN-Kubernetes delivers industry-leading performance, functionality, and efficiency. Running OVN-Kubernetes on the DPU reserves host CPUs exclusively for workloads, maximizing system resources.

Deploying and managing DPUs and their associated DOCA services—especially at scale—can be challenging. Without a provisioning and orchestration system, the complexity of managing the DPU lifecycle, deploying DOCA services, and providing the necessary network configuration on the DPU to redirect the network traffic via those services (service function chaining, or SFC) becomes a significant burden for cluster and system administrators; which is where the DOCA Platform Framework (DPF) comes into play.

DPF simplifies DPU management by providing orchestration through a Kubernetes API. It handles the provisioning and lifecycle management of DPUs, orchestrates specialized DPU services, and automates tasks such as service function chaining (SFC). This ensures seamless deployment of DOCA services like OVN-Kubernetes and HBN, allowing traffic to be efficiently offloaded and routed through HBN's data plane.

With DPF, users can efficiently manage and scale DPUs within their clusters while automating critical processes. DPF orchestrates the deployment of OVN-Kubernetes and HBN, optimizing performance with features such as offloaded OVN-Kubernetes CNI and accelerated traffic routing through HBN.

This RDG provides a comprehensive, practical example of installing the DPF system on a Kubernetes cluster. It also demonstrates performance optimizations, including Jumbo frame implementation, with results validated through standard RDMA and TCP workload tests.

© Copyright 2025, NVIDIA. Last updated on Jul 10, 2025.