How to start a successful journey with Open Platform for NFV

Sponsored article

One of the key goals of NFV is to decouple network functions from dedicated and proprietary platforms so that carriers can benefit from higher efficiencies, reduced costs, and accelerated time-to-market for new services. The ability to deploy Virtualized Network Functions (VNFs) on commodity hardware and open source platform enables operators to more easily use best-of-breed vendors without vendor lock-in.

Since November 2012, the ETSI NFV ISG (European Telecommunication Standards Institute, Network Functions Virtualization, Industry Standards Group) has taken a lead role in defining reference architectures and integration reference points (APIs) for NFV. Early on, this group began exploring ways of collaborating with other industry groups to accelerate the adoption of NFV. As a result, In September 2014, the Linux Foundation announced the founding of the Open Platform for NFV Project (OPNFV). Red Hat was one of the founders and as a Platinum member we’ve been one of the main contributors ever since. The community driving this project began working closely with ETSI’s NFV ISG, among others, to drive consistent implementation of standards for an open NFV reference platform. Today, OPNFV is an open source reference platform to accelerate the introduction of new NFV products and services.

OPNFV has been mainly focused on building NFV Infrastructure (NFVI) and Virtualized Infrastructure Management (VIM) by integrating components from upstream projects such as Linux, OpenStack, Ceph Storage, KVM hypervisor, QEMU (for hardware environment emulation), libvirt (for hypervisor management), DPDK (Data Plane Development Kit for accelerating packet processing), OVS (Open vSwitch for packet switching functionality), and OpenDayLight plug-in (SDN control). These components, along with application programmable interfaces (APIs) to other NFV elements form the basic infrastructure required for Virtualized Network Functions (VNF) and Management and Network Orchestration (MANO) components. Recently, OPNFV has also started working with open source MANO projects, as a way to accelerate the development of standards of important NFV interfaces, such as VNF onboarding.

Building a truly open NFV platform

The OPNFV is an open source reference platform that is defined by several individual software components that need to be put together in a comprehensive and cohesive manner. To be able to put together an integrated NFV software platform that is consistent, high quality and high performance, the following factors need to be taken into account at a minimum:

A complete understanding of all separate open source projects that are to be integrated from an engineering point of view.
The ability to adopt different community releases at the right time and to coordinate the release cadences of these projects.
The experience and know-how to influence the communities in the right way to have needed features and functionality at the right time.
The capability to coordinate the release schedules of these projects.
The resources to integrate and test all the components together and perform quality assurance.

At Red Hat, this is exactly what we have done for our customers in providing them an NFV platform. We’ve integrated Red Hat Enterprise Linux, Red Hat OpenStack Platform and Open vSwitch based on DPDK into our customers’ NFV systems for their needs including SDN-enabled Virtual RAN, Packet Core and Value Added Services (VAS). To achieve required performance levels on networking, we used high-efficiency and high-performance extensions to the virtualization and networking layers, such as Virtual CPU (vCPU) pinning, memory hugepages support and DPDK-accelerated Open vSwitch.

Open source is the key to your success

But we didn’t stop there. We believe adding high availability, performance and security into these technologies and integrating them upstream is critical to building effective open source products. Open source innovation works best if everyone contributes to the main source code of the project. Adding to open source code without contributing that code back to the community creates new versions that may become incompatible with the main version over time. This locks you to your vendor and requires costly migrations to regain interoperability and get access to new features.

At Red Hat, we contribute all of our code developments upstream first, then we build distributions from the upstream code into our core products, ensuring that each one is truly open source. This enables our customers to achieve more cost-effective, sustainable innovation without vendor forking, branching or lock-in. Our customers can use the software knowing that it is truly open source while enjoying many benefits that we offer, including hardened and tested distributions, managed upgrades and updates, vulnerability management, and world-class support.

OPNFV is a great starting point for building a truly open NFV platform, but because of its complexity, ongoing improvements and many moving parts, it is essential to address all of the technical and administrative challenges identified above. This is the only way to ensure a consistent, high-quality, high-performance and high-security NFV software platform. Any service provider looking to deploy NFV should either take it upon themselves to address these challenges or should consider getting their distribution from a supplier like Red Hat that does it all for them. At Red Hat, we know all the required software pieces for creating a consistent, high-quality and high-performance OPNFV, have the experience of integrating them together, and a track record for high standards of hardening, testing and support. This is the right way to start your successful journey with NFV.

Ali Kafel , Technical Product Marketing Lead , Telecom, Red Hat