Open source initiatives kicked off by AT&T are gaining significant influence in the industry, from ONAP (Open Network Automation Protocol) in the virtualized network orchestration space; to the newly announced dNOS operating system for telco white boxes; to xRAN, which seeks to define a “software-based, extensible RAN and to standardize critical elements of the x-RAN architecture”.
If these succeed in becoming de facto standards, they could help AT&T drive the transformation of its own networks and cost base, and those of the telco sector as a whole, while retaining control of the value chain at the expense of proprietary vendors on one hand, and web players on the other.
Now, xRAN.org is preparing its first specifications, hard on the heels of the first ONAP release, though like ONAP, it will have to work hard to address criticisms that it will be unable to deliver a platform with the carrier-class robustness of traditional networks.
Nonetheless, the open source group says it will release its first specs this month or very early next year, based on work to decouple the RAN control plane from the user plane. It has been working first on three areas which will help to transform the RAN:
- standardized northbound and southbound interfaces to make it easier to deploy services and achieve multivendor networks, in a decoupled, software-defined environment.
- implementation of RAN software and network functions on off-the-shelf hardware,including the modular xRAN LTE eNode B base station software.
- and the flexibility for operators to run software functions at the macro cell site or at the edge of the network, depending on availability of fronthaul links.
Although the initiative started in AT&T, which has contributed technology, there will also be inputs from other members, which include virtualized packet core supplier Mavenir (which is also a leading light in the open source CORD group), as well as operators including Verizon, Deutsche Telekom, SK Telecom and Telstra. The group is trying to avoid third party intellectual property in its specs, presumably to maximize the cost and innovation benefits of open source while protecting adopters from exposure to patent fees down the road.
Mavenir says it has the specs already running and in trials with four large carriers. According to Mavenir’s John Baker, the next piece of work will be the operation and management (OAM) interface, to allow for multivendor interoperability between remote radio heads and baseband units, a key enabler of Cloud-RAN, in which the radio heads are managed by a centralized and virtualized controller in the cloud.
Baker says xRAN is working with five or six RRH suppliers and he expects commercial hardware to be available in the first half of 2018, adding to the challenges to the traditional proprietary, end-to-end model for the RAN, which has been dominated by a shrinking group of large vendors.
However, it will not be easy for these open source groups to convince operators to trust their key networks to the new platforms. It is one thing for telcos to put pressure on their established suppliers by banging the drum for open initiatives and for groups like Facebook Telecoms Infra Project (TIP). But it is quite another for them to risk deploying critical functions and services on a brand new architecture. Hence the mutterings about the initial ONAP release, with some operators saying it would take several more versions before they would regard the software as carrier-class.
And even in the open world, there are other challengers to the AT&T-driven developments, notably from the more traditional ETSI camp. ETSI’s Open Source Mano (OSM) is a competitor to ONAP in the critical area of virtualized network management and orchestration (MANO), and has recently been putting more pressure on its upstart challenger.
And it has recently expanded its remit, setting up new groups focused on zero-touch networking and automation. A brand new group is called Zero Touch Network Service Management (ZT-NSM), which aims to tie the MANO into the rest of the management layer, and is largely based on code from Deutsche Telekom. This will be separate from OSM, but shows ETSI’s various groups starting to stitch together a broad platform driven, like ONAP, by operators’ requirements and inhouse developments (OSM draws heavily on Telefonica code, for instance).