Building an Extensible Open vSwitch Datapath

Abstract

The virtual switch is the cornerstone of the today's virtualized data center. As all traffic to and from virtual machines or containers must pass through a vSwitch, it is the ideal location for network configuration and policy enforcement. The bulk of Open vSwitch functionality is platform-agnostic and portable. However the datapath, which touches every packet, is unique to each supported platform. Maintaining each datapath requires duplicated effort and the result has been inconsistent support of features across platforms. Even on a single platform, the features supported by a particular kernel version can vary. Further, datapath functionality must be broadly useful which prevents having application-specific features in the fast path. eBPF, extended Berkeley Packet Filter, enables userspace applications to customize and extend the Linux kernel's functionality. It provides flexible platform abstractions for network functions, and is being ported to a variety of platforms. This paper describes the design, implementation, and evaluation of an eBPF-based extensible OVS datapath. The eBPF OVS datapath delivers the equivalent functionality of the existing OVS kernel datapath, while significantly reducing development pain points around maintainability and extensibility. We demonstrate that these benefits don't necessarily have a trade off in regards to performance, with the eBPFbased datapath showing negligible overhead compared to the existing kernel datapath.

Comparing the performance of state-of-the-art software switches for NFV

“…ESwitch [45] employs template-based code generation to optimize Open-Flow pipeline. OVS eBPF [46] presents a kernel-based datapath optimization for Open vSwitch with extended Berkeley Packet Filter (eBPF). These solutions cannot be included for direct comparison as their code is not currently available.…”

Beyond socket options: making the linux TCP stack truly extensible

“…Several networking use cases already leverage eBPF. For example, XDP uses it for fast packet processing [37], IPv6 Segment Routing uses it to support network programming [67] and it improves the extensibility of Open vSwitch [61]. A key benefit of the in-kernel eBPF virtual machine is that each eBPF bytecode is provably verified before being injected inside the kernel.…”

Cadeia-Aberta: Arquitetura para SFC em Kernel Usando eBPF

“…InKev [Ahmed et al 2018] habilita a execução de programas no plano de dados de redes virtuais, com foco em redes de centros de dados. Tu et al [Tu et al 2017] descrevem o projeto, implementação e avaliação de um plano de dados extensível para OvS, baseado em eBPF. Para permitir que o protocolo OpenFlow suporte protocolos diversos, Jouet et al [Jouet et al 2015] definem uma extensão para o protocolo, chamada OpenFlow Extended Match field (OXM), para instalar bytecode BPF.…”