Ericsson will not be content to let Nokia and Samsung make all the running in big-vendor O-RAN. Earlier this month, the head of the company’s networking business, Fredrik Jejdling, said Ericsson would take a major role in the emerging open RAN market – not by being first with a compliant product, but by being the first to deliver an offering that can match, or improve on, the performance of existing dedicated systems.
He told an online investor event that Ericsson believed in open ecosystems and already had a “prominent and impactful role” in the O-RAN Alliance, but added: “We need to put a realistic view on how and when it will be introduced into the networks to deliver performance and cost benefits … Off-the-shelf hardware will not [currently] be able to achieve the same cost benefit as our current integrated solution.”
But he conceded: “Open RAN specifications will improve and at the point of time when open RAN-specified radios, basebands and configurations have the cost and performance benefits to the level of the integrated solutions we provide today, we need to be the first player providing those in parallel to our current portfolio.”
This is a common Ericsson approach to the opening up of the RAN, and echoes that of Cisco and others in the enterprise and wide area networks. It involves voicing support for open platforms, while pointing out their shortcomings (usually quite fairly) compared to vertically integrated, dedicated systems with years of engineering resources and deployment expertise behind them. The subliminal message is: by all means deploy open architectures for secondary networks – indoor small cells, rural extension and so on – but introduce them too hastily to the primary network at your peril.
This resonates, of course, with operators which can lose revenue and market share overnight because of an outage, and which are currently hounded by companies testing their speeds and quality of service relative to competitors’. It also speaks to the more conservative factions within many operators – while visionaries and CFOs may embrace the economics of open cloud networks, those whose skills and power rest on large engineering organizations with almost mystical skills find it hard to concede that an open platform could achieve the same results.
For now, they don’t in most scenarios, but majors like Nokia and AT&T are working hard to enhance frameworks like O-RAN, and the Telecom Infra Project is making good progress on practical deployment blueprints, plus there are the innovations and determination of a crowd of smaller providers. So Ericsson cannot be complacent, but it will play the caution card for as long as possible.
This was clear last year when it announced a cooperation with Nvidia to develop an experimental vRAN platform with the chip provider’s GPU (graphical processing unit) architecture supporting the accelerations rather than specialized, single-purpose chips. That saw Ericsson putting resources and public support behind a development that could shake up the ecosystem and significantly improve the cost of advanced vRAN – but it did so with significant notes of caution, sounded loudly by Thomas Noren, Ericsson’s head of 5G commercialization.
He said Ericsson’s existing baseband, running on specialized silicon, could already handle the “incredibly computer intensive” software needed to support a demanding 5G workload such as dynamic spectrum sharing (DSS). In terms of replacing that with the Nvidia platform, he was merely “open to the idea”. He said: “We will explore if we can develop a distributed baseband product (DU) with Nvidia GPU. We don’t have any committed product plans, but we think this is a very interesting idea.”
So Jejdling has been continuing with the familiar strategy – protect the traditional platforms, with their lock-ins and all their associated integration and support revenues, for as long as possible, but invest in being able to seize the lead at any stage when the new open platforms go mainstream.
However, he appeared to dismiss any prospect of joining the Open RAN Policy Coalition, which was set up recently to lobby US regulators and lawmakers, in particular, to back open RAN architectures by giving them preference (or even making them mandatory) for government projects and grants, and potentially any national infrastructure.
That would achieve the double whammy, for some US politicians, of boosting new vendors which are primarily US-based, resurrecting an American RAN industry; and providing operators with a broader and more price-competitive alternative to Huawei. The latter point is moot in the USA, where Chinese equipment is barred from 5G roll-outs, but could help to persuade governments elsewhere to adopt US-style sanctions against Huawei.
But while Nokia has joined the Coalition, along with most of the O-RAN start-ups such as Parallel Wireless and Mavenir, Jejdling said: “We believe these types of forums should not define the types of technology [used]. That should be the industry in line with policy recommendations. Whether that’s virtualized RAN, whether that’s cloud-based RAN, whether that’s open RAN or an integrated solution. The industry should evaluate based on the technology.”