The past 15 years have been ones of a shrinking vendor base in mobile networks. Operators whose governments wish to restrict Chinese equipment in 5G build-outs complain bitterly that they will be left with a choice of only two global and long-standing suppliers, Nokia and Ericsson, with Samsung as a possible counterweight.
But the argument is over-simplified. In some respects, as the 5G network transforms to new virtualized and disaggregated architectures, the supply chain has not looked as diverse for many years.
This is not really about the wave of specialist vendors pinning their colors to the O-RAN mast. Initiatives like the O-RAN Alliance and Telecom Infra Project are greatly to be welcomed as they show the way to a truly open, multivendor and affordable set of foundations for 5G. If the hijacking of O-RAN in the interests of geopolitics does not spark a backlash, it (or a similar successor) should have a strong influence on making the ecosystem broader and more diverse, so that cellular equipment can support the full variety of use cases and industries, rather as WiFi does.
But for now, new architectures are mainly for enterprise, private, rural and small cell scenarios, in our view – greenfield and reasonably small-scale deployments which are separate from the main macro networks and do not present a risk to the core applications and user experience. That is an attractive market for start-ups or specialized vendors, and will allow some of them either to build a healthy business, or attract the attentions of an acquirer looking for mobile network expertise – perhaps Cisco will try the RAN again, though its sale of its SON business may suggest not; perhaps other webscalers will follow Microsoft’s recent purchase of Affirmed Networks and Metaswitch.
But when it comes to the national-scale macro networks, the dynamics are different, as are the new entrants to the supply chain. Understandably, non-greenfield operators tread cautiously when it comes to disrupting and transforming the main networks on which nearly all their revenues rely. So here, the choice of vendors is opening up in a different way.
It remains true that nobody understands how to design and optimize a macrocell radio and antenna like the big four suppliers. But the heart of the 5G network is all about cloud infrastructure, virtual network functions and software-defined networking (SDN), approaches that are far more mature in the enterprise, data center and webscale worlds than in telecoms. That has always held out the potential to allow vendors from these markets to cross the divide into the telecoms network area. And now those enterprise players are making their moves, in anticipation of a wave of deployment of cloud-native, SDN-enabled 5G cores and even RANs over the coming few years.
Oracle, IBM, Cisco, Juniper, HPE, IBM, Dell and others are all, in their different ways, refining their 5G strategy. All of them have telecoms businesses, providing data center hardware and back office applications, or professional services, to telcos; or in the case of Cisco and Juniper, switches and routers. But now they can, for the first time, ride the coat-tails of virtualization and SDN to move into the network itself.
They need to firm up their propositions quickly, because Nokia, Ericsson and Huawei have all been developing their own offerings, in a bid to keep the drawbridge up against the data center barbarians. They have not just virtualized the network functions, but developed their own cloud hardware to support them. Nokia’s homegrown processors and core routers are challenged only by Cisco, and it is targeting them at hyperscalers as much as telcos.
The established vendors may subscribe to open cloud standards like OpenStack and Kubernetes, or to emerging open network specs like ONAP and ORAN, but that does not stop them developing a vertically integrated hardware/software platform. The lock-in of an end-to-end solution might be more voluntary than current ones, but it can still be hard for MNOs to escape if they have invested in every piece of a network from the same supplier.
So the data center players will need to convince operators quickly that they not only bring the advantages of their long experience with virtualization, SDN and cloud; but they will also support full openness, and will have the partnerships in place to deliver the actual radio functionality.
That is where Samsung and NEC have an interesting opportunity – and some of the smaller O-RAN companies like Airspan may too. They are too small in the network infrastructure market to offer end-to-end solutions but they have high quality RAN solutions, they support open interfaces, and they have a good track record with advanced operators, even if only in small numbers (NEC with NTT Docomo and now Rakuten, Samsung with SK Telecom and now the US fixed wireless deployments, Airspan with Sprint, Reliance Jio and Rakuten).
So they could be ideal partners for cloud-based network vendors. NEC is a particularly interesting case. It has never succeeded outside Japan in the macro RAN, though it has small cell and backhaul businesses, but it is well-established in the OSS layer through its acquisition of Netcracker. Now it is working with Rakuten to develop and market a 5G radio and core round the world.
With the UK apparently interested in NEC as a possible addition to the 5G supply chain (see 5G Watch below), easing the pain, for operators, of excluding Huawei, we are likely to see more of the Japanese company in future – and an intensifying battle between the stalwarts of the data center to become the new giants of the mobile network ecosystem.
The consolidation of the mobile network supply chain:
The mid-2000s were characterized by M&A – Alcatel merging with Lucent, Nokia and Siemens forming a networking joint venture, Ericsson buying Marconi Networks. Then at the turn of the decade, old names started disappearing altogether – Nortel collapsed, Motorola was broken up and its cellular business bought by Nokia. And in 2016, the Alcatel name was gone too, when Alcatel-Lucent was acquired by Nokia.
Through all this consolidation, of course, there was one major element of diversification of the supply chain, as China’s Huawei and ZTE became more and more successful in mobile networks. The rise of Huawei, in particular, has been an important factor in the current 5G-centric battles between the USA and China over trade, cybersecurity, intellectual property and the ‘5G race’.