Your browser is not supported. Please update it.

31 January 2018

Linux Foundation fights fragmentation with network umbrella

Open source platforms are becoming more and more fundamental to the new telecoms network architecture, raising exciting prospects for a more democratic ecosystem and rapid innovation. But open source also comes with the risk of fragmentation, which has already been seen in industry splits over different approaches to management and orchestration (MANO) in virtualized networks.

A large number of open projects has emerged in the areas of virtualization, software-defined networking (SDN), MANO and even telecoms hardware initiatives like Facebook’s OpenCellular. Many of the most prominent are hosted by the Linux Foundation, and that increasingly powerful force in the telecoms landscape has announced that it will be combining six major projects under a common “horizontal umbrella” called the LF Network Fund (LFN).

This is a bid to rationalize activities and reduce fragmentation and market confusion – and the resulting threat that operators will lack the confidence, despite the allure of open platforms, to tear themselves away from their familiar vendors, standards bodies and closed platforms. It will also help to reduce the variety of skills and resources telcos will need to support all the projects, and will charge a single fee to belong to LFN.

The LFN will bring together six projects at various stages of maturity. They are:

  • ONAP (Open Network Automation Protocol)
  • OPNFV (Open Platform for Network Functions Virtualization)
  • OpenDaylight
  • io (Fast Data.input/output)
  • PNDA (Platform for Network Data Analytics)
  • SNAS (Streaming Network Analytics System)

The open source platform has already adopted the umbrella approach with its existing Cloud-Native Foundation. In these structures, individual projects retain their independence but the foundation provides mechanisms for communication and sharing, and enables “avenues for greater collaboration” between them. This is achieved through two LFN bodies, a technical advisory council and marketing advisory council.

Arpit Joshipura, general manager of networking and orchestration for the Linux Foundation, told LightReading: “Over the last five years, open source has picked up steam significantly and as specific use cases or problems arose, multiple single projects were created, and they have been deployed and have had momentum. Now there is an architectural discussion going on across projects and there is a need for looking at all these components from a solution perspective.”

Joshipura gave the example of the on-boarding of virtual network functions (VNFs), where ONAP and OPNFV both have activities. ONAP has created a set of guidelines and software development kits (SDKs) but OPNFV goes a step further, “allowing VNF vendors to look at the entire stack to get to a to a certification level,” he said. “Now if we combine the two, you have everything … The full lifecycle management of VNFs is synchronized.”

More umbrellas will be created over time, the Foundation says, with the aim of promoting harmonization and creating a core set of projects which are fundamental to SDN, virtualization and the transformation of telecoms networks. Other telecoms-related projects it hosts include Open vSwitch, OpenSwitch, IO Visor, ON.Lab, CORD, ONOS (those last three closely related) and XRAN.

However, it is treading a delicate line between addressing fears of fragmentation, and appearing to behave like a traditional standards body. It insists that projects will keep their independence and that no project has to join the LFN – each one will decide if and when to take part.

Of course, this difficult balance epitomizes that which operators will have to draw too – between the certainties and predictability of traditional standards and platforms; and the opportunities in open source to support open multivendor platforms, and to reduce long term total cost of ownership (though not upfront cost – telcos have been clear that open source technologies like Openstack currently require significant outlay on consultants and integrators, or on inhouse experts).

Mansoor Hanif, director of BT’s converged networks research lab, said he liked the fast pace of open source, but had needed “a whole new generation of network engineers”, and was concerned about interoperability.

Some operators are downright hostile – Shahar Steiff, an assistant VP at PCCW, told last year’s TMForum Live event in France that open source “only provides half of the things we need – the code but not the information model or standards. Yes, it is faster than proprietary code, but with a standard I don’t care if it is open source or proprietary code.” And he added that, while open source was ‘free’ upfront, it cost a great deal to implement effectively in an operator’s systems.

So there are risks as well as rewards in the open process, and most players are looking towards a hybrid approach where the speed and broad-based innovation of open source can be tempered by the clear direction and strong quality control of a formal standards process.

And the Linux Foundation itself confronted the dilemma head-on last May when it published a white paper outlining how different open source initiatives, and standards bodies, could work together to avoid duplication or fragmentation. It called for telecoms standards bodies to work together with the open initiatives to accelerate progress and avoid damaging splits or fragmentation.

Multivendor interoperability and automation across multiple NFV/SDN efforts will require close coordination, said the white paper, as well as recognition, on both sides, of what each does best.

The paper, entitled ‘Harmonizing Open Source and Standards in the Telecom World’, points out legal and intellectual property challenges and ways to work together, but also builds on a unified architecture which Joshipura presented at last year’s Open Network Summit. That aligned various open source groups with existing standards bodies and pointed out areas where they were overlapping or duplicating effort.

As Joshipura put it: “There’s a place for standards and there’s a place for open source, and the two of them can be the best of friends.”

He added: “As a neutral party, we would like to facilitate those conversations, but some of that may take place in the standards groups themselves. The first step is project by project, standard by standard. We will look at introduction goals, what are complementary things. We agree on the end results but the paths are different on how we get there and when the paths are different, it’s important to get terminology aligned and processes aligned and start the dialog for the actual architecture and the integration. Open API does not mean the same thing when I say it as when a vendor says it or when another open source project says it.”

Of course, another step towards harmonization comes when sufficient weight of major operators and vendors actively supports a certain platform or approach. ONAP, which was originally based around AT&T’s ECOMP MANO system, with inputs from China Mobile, scored a significant coup at the start of the year when Verizon also signed up as a platinum member.

But there are still other MANO initiatives which may limit ONAP’s power, or at least spread uncertainty or delay deployment decisions among MNOs. The most prominent is ETSI’s Open Source MANO (OSM), to which Telefonica has been a prominent contributor.

Last summer, there were reports that the Spanish telco might join ONAP too, when its planning and technology director, Javier Gavilan, said in an interview: “OSM cannot be compared with ONAP because the scope of ONAP is bigger and OSM is only a small part of the Telefonica transformation project. We are transforming our full stack and this is something we are doing and it could be a part of ONAP.”

Verizon has also said that it sees ONAP has being broader, providing a full service management platform, while OSM is mainly an orchestrator for virtual network functions (VNFs).

Telefonica has not yet joined ONAP, but if it does end up belonging to both groups, it could be a catalyst for convergence, which would help address one of the many sources of fragmentation in the immature world of carrier SDN/NFV. Gavilan said last August that he had been in contact with AT&T about aligning ONAP and OSM efforts to develop common information models and processes. He said: “The industry has two different initiatives running in parallel with a lot of common points and the idea is to align them as much as possible.”