- As Open RAN is all about flexibility and agility, an inability to nail its lifetime costs is hardly surprising
- The TelecomTV Open RAN Summit revealed a mostly confident ecosystem, an abundance of working APIs and a new willingness to collaborate
Despite all the progress, some in the industry still have the jitters over Open RAN. Will it be worth all the effort? And the risk? One reason for these doubts is that, so far, there are only a handful of full-blooded Open RAN implementations – usually in greenfield networks – and, as a result, there is scant concrete evidence that Open RAN brings with it much financial advantage in return for all the technology and deployment risk.
Of course, on paper and in the abstract, a good case can be made. Open RAN is designed to give cellular operators a commercial leg-up by stimulating the entry of new vendors into a market that is currently dominated (in most of the world) by just two, thus allowing telcos to build differentiated networks by exercising choice when it comes to the specific network elements they deploy. But whether this and other advantages can make up for the extra costs (and unknowns) involved in this radical change of approach is still uncertain.
At present, many operators appear to be biding their time and waiting for more evidence that Open RAN is the way to go before they dip their toes. Just as important, they’re asking: is it the way to go right now? Why not wait for the smoke to clear and the post-Covid, post- supply chain turmoil, post-Ukraine war, global economy to start driving network growth – especially for 5G – before we start serious commercial rollouts?
Given all this, it’s hardly surprising that when we asked delegates at TelecomTV’s recent Open RAN Summit what the industry should regard as the most important area of focus for Open RAN over the next 12 months, the answer came back loud and clear – determining total cost of ownership (TCO) topped the poll. The TCO question also cropped up several times throughout the summit and our expert panel members imparted some excellent advice.
Total cost of ownership
One fundamental problem, as we discussed in the first summit report, Calculating the impact of Open RAN, is that Open RAN is, above all, a way of hooking up the benefits of disaggregation in terms of agility and the ability to meet the needs of specific customers, or groups of customers, according to Appledore Research analyst Francis Haysom. “A lot of the TCO [improvement] will not just be about minimising costs today, but about providing an opportunity to do something more: to be more relevant to particular enterprises or use cases,” he noted.
For instance, Open RAN equipment choices may enable an operator to shape commercial offers around ultra-high performance, say, with low-latency high-speed services; another may opt for mass connectivity for ultra-low-speed IoT with a cellular infrastructure able to handle tens of thousands of low-speed connections. The technology used in the RAN will define how successful and cost effective (for the operator) such an approach is.
Open time frame for Open RAN
What’s the time frame for open RAN? This was an allied question to the extent that calculating a TCO might require a timeframe within which to amortise equipment costs.
However, according to our experts, there is no timeframe with Open RAN.
It’s open-ended, said Andy Dunkin, OpenRAN radio frequency and digital platform development manager at Vodafone. The idea, he added, is to foster an ecosystem that supports all the generations of mobile technology, partly in recognition of the fact that they have proven to be unexpectedly long lived and, therefore, difficult to “sunset”: Many 2G applications and handsets, for instance, are still alive and kicking today and, while 3G was a qualified failure, 4G looks like it will persist well into the future to rub shoulders with 6G. The fact that sustainability now has a high priority means that, if anything, device lifespans will continue to lengthen as users keep them longer and are encouraged to do so, rather than be persuaded by shiny object marketing to buy the latest products.
He is supported in this opinion by Volker Ricker, director and product line manager of Open RAN at CommScope, who said it doesn’t matter which network technology is being considered – whether it is 2G, 3G, 4G or 5G – all must be included with Open RAN for the foreseeable future as markets demand it.
Old is beginning to be good
An interesting development in the phone market, which may validate the Open RAN approach to broad support for the ‘Gs’, is the way owning an old or retro phone is slowly but surely becoming something to be admired rather than scorned, partly for trendy sustainability reasons, but partly too because of a distinct turning away from the fully loaded smartphone concept by a proportion of users who are tired of its constant demand for attention and have opted for a so-called digital detox, often foregoing engagement with social media as a cold-turkey exercise.
Telcos may be able to avoid their own version of cold turkey when it comes to giving up some of their old tried-and-true approaches and expectations to become successful adopters of Open RAN, though.
According to Rimma Iontel, chief architect of Red Hat’s TME technology, strategy and execution office, telcos can take little steps and backtrack if a move doesn’t meet expectations. “You can even test something out on another infrastructure ,” she says. “You don’t have to roll out to the whole network, just a little section.” This dynamic, she says, is what people need to internalise.
That’s a big ask for many in an industry that is historically used to long testing cycles for new technologies and the overseeing of their careful integration before they are let loose in the network. Then, once systems are in place and working, some operators are still loath to make software changes. This mindset is the polar opposite to that required for cloud-native software and the use of techniques such as continuous integration/continuous delivery (or deployment) (CI/CD) and DevOps. (CI/CD is a set of practices for rapidly and reliably delivering code changes across a network while DevOps is designed to have development and operations teams work together to streamline product development).
So when it comes to nailing down TCO, or even just providing a framework for its calculation, there are so many variables and unknowns that those in search of hard numbers ‘to take to the board’ will find them difficult to find.
So where are we?
Our panellists and interviewees for the TelecomTV Open RAN Summit, for the most part, remain bullish on the prospects for Open RAN. According to Vodafone’s Andy Dunkin, the ecosystem is developing nicely. “Radio units are very well supported. We see a healthy catalogue of single-band radios, which is where everybody’s starting. But now we’re pushing our vendors for dual- and triple-band macro radios to support those initial deployments,” he said. “Plus there is new silicon coming through to support those products – ready for testing at the end of this year and ready for commercial deployment in 2023.”
He added that Vodafone is developing further reference designs for contract manufacturers to pick up to support Vodafone’s own white-box radios.
Arguably the most successful Open RAN vendor so far is Mavenir, which supplies the software for Dish Network’s ambitious Open RAN deployment in the US. According to Stefano Cantarelli, CMO at Mavenir, in a conversation with Ray Le Maistre, “Dish proves that a big rollout of Open RAN can be accomplished now,” and he claims that Dish will become the reference architecture, built on standalone 5G. Non-standalone won’t cut it, he said: “It’s just providing an extension to 4G. Standalone becomes the key focus for all the cloud-native, automation and elasticity goals for the next generation of mobile networks. All the advanced 5G applications require standalone,” he maintained, “and all the real differentiation is happening in the standalone space.”
Supplier diversity: Yes. Tech diversity: Not so much
Also recorded in an executive interview for the Open RAN Summit by TelecomTV’s deputy editor, Yanitsa Boyadzhieva, was Neil McRae, chief architect at BT. He expects that BT’s trials will show that the RAN intelligent controller (RIC) is likely to be particularly valuable in future, but he cautioned that, so far, there are few signs of the much-vaunted differentiation that was supposed to be the hallmark of Open RAN. “I think it’s given us more people to buy from, but I’m not sure that the technology is wildly different,” he says. “It’s great that there’s more choice at one level, but I think there’s much more the industry has to do to improve the availability of some of the key components.”