Supplier diversification is a primary goal of the Open RAN ecosystem and the Telecom Infra Project
MADRID—Vodafone has been a driving force in cultivating the Open RAN ecosystem as it works towards the goal of using disaggregated radio systems in 30% of its European networks by 2030. With nearly two dozen owned and operated networks around the world, and 14 of those in Europe, this will clearly be a major undertaking. At the Telecom Infra Project’s FYUZ event, Vodafone’s Yago Tenorio, director of network architecture and TIP’s board chairman, in his opening remarks ran through a number of Open RAN-related press releases Vodafone put out today, saving for last what he termed his “favorite announcement. … We agreed that Nokia is going to deliver a fully-compliant Open RAN system—the first incumbent supplier that has actually fully embraced Open RAN.”
In addition to Vodafone’s stated goal for Open RAN, Deutsche Telekom, Orange, Telefonica and Telecom Italia have also committed to deploying Open RAN at scale. These commitments are meant to signal to Open RAN (and incumbent) vendors that multi-vendor RAN is happening, so get on board. But today’s announcement, taken in the context of Nokia’s historical combativeness towards and criticism of Open RAN, comes across as a bit mixed.
For example, speaking this summer at RCR Wireless News’ Open RAN European Forum, Adrian Hazon, Nokia’s Vice President of global product sales, questioned the business case for Open RAN in both greenfield and brownfield environments. The question, he said, is whether “O-RAN can really deliver the cost savings and the equivalent performance that’s expected. That’s the key challenge to this, and it’s not close to being proven yet.” To the supplier diversification angle that’s a central tenet of the Open RAN ecosystem push, “More competition doesn’t necessarily lower the cost or improve the performance,” he said.
In a piece from Light Reading’s Iain Morris, Nokia President of Mobile Networks Tommi Uitto shared similar thoughts. “There is an assumption that more suppliers is more competition and prices come down, which I am not sure will happen. Is the cake big enough for 10 different suppliers? There used to be 10 different base station suppliers, but we have gravitated to a smaller number.” He continued in that interview: “Sometimes people think Open RAN will be cheap. No, it’s not. There’s nothing that would make it inherently cheap because it is 3GPP-compliant basestations, but additionally it has the particular software-made open interfaces. … Actually, in some early versions from the Open RAN challengers, they even had to use FPGAs to make those open interfaces, just as we did in our first 5G products, and that certainly does not make your product cheap.”
So what has changed that Nokia has now “fully embraced” Open RAN? I asked just that in an email to Nokia this morning and will update this story with a response as possible. Worth noting that in the relevant Vodafone press release titled, “Vodafone and Nokia partner to advance Open RAN ecosystem in Europe,” no Nokia executive joined Vodafone’s chief technology officer or Marvell’s president of products and technologies in providing comment. Perhaps also worth noting that Nokia doesn’t have a stand here at FYUZ; they do, however, have a stand at Vodafone’s Arch Summit which starts tomorrow in Luxembourg, according to a source on the ground at Luxexpo.
As to the announcement, Vodafone said it will work with Nokia on a fully-compliant O-RAN solution bringing Nokia’s ReefShark SoC, which was co-developed with Marvell, to commercial-off-the-shelf servers in order to “enable the Open RAN system to reach functionality and performance parity with traditional mobile radio networks.” The statement continues to note, “This partnership with Nokia will help Vodafone meet its public target of having 30% of its European networks running on Open RAN by 2030, as well as boost the EU’s global technology leadership in digital infrastructure.” The next step is to demo an Open RAN baseband “early next year.”
What about the Ericsson of it all? In a panel here at FYUZ made up of representatives from NTT DoCoMo, Omantel, Orange, and Telefonica, Tenorio, serving as moderator, asked the group, “Will 5G be the last generation of the incumbent architecture?” Ideally we’ll cover the responses in more detail at some point, but for the purposes of this, we head to a reply to my Twitter post sharing the question. In response, Gabriel Foglander, a strategic product manager for Ericsson’s Cloud RAN portfolio, wrote, “Asked at an event promoting…yeah, that’s what I thought! Meanwhile they and approx 99% of users worldwide continue to tweet/surf/call happily on said architecture.”
In a conversation at the show with Jefferson Wang, Accenture’s senior managing director of the Global 5G and Networks practice, we talked a bit generally around Open RAN as a vector for supplier diversification, the economic reality that RAN is a scale business, and what apparent buy-in from a large incumbent means for smaller firms looking to take share in the market. “There are definitely many clients and operators out there that are very comfortable with the incumbents,” he said. “They feel confident in what they can do because it’s an entire ecosystem. … There’s this innate comfort with it.”
Then Wang shifted the frame a bit. “I guess the question is, will that either accelerate or stifle the whole movement” if incumbent vendors embrace Open RAN?
For Mavenir’s John Baker, senior vice president of business development, that Vodafone is working on that 30% goal with Nokia is antithetical to supplier diversification, a broad, long-standing objective of the Open RAN movement and an explicit aim of TIP. From the TIP website: “We seek to expand the supply chain and drive innovation across the entire telecom landscape. That means collectively designing, building, and testing technologies that are more efficient and interoperable across the whole product lifecycle. TIP aligns a diversity of skills and creates economies of scale to accelerate commercial solutions.”
“There’s no vendor diversification which was actually the stated goal,” Baker said. “At the end of the day, the operator will suffer and, to put it back, the consumer will suffer.”
Back to the “economies of scale” part of the TIP language. This is something Baker has publicly discussed for some time–for non-incumbent Open RAN vendors to be competitive, they need to be able to buy components at price points similar to those larger incumbents get given the volume they purchase. “The reality is that the small companies are funding the discounts of the large companies,” Baker said.
“If the TIP vendors are going to carry on with the same behavior, there needs to be a way to defeat it. In reality what should happen is that any chips that are being used in volume by the total Open RAN community should …have an Open RAN price on them. All Open RAN members get the leverage of one another.”
Counterpoint from Joel Brand, Marvell senior director of product marketing, during a panel Baker also sat on. “When you’re the size of Nokia, you can get Nokia’s pricing.”
One could certainly make a case that Vodafone working with Nokia to put the latter’s SoC into a COTS server for deployment in the former’s network isn’t materially different from what Mavenir and other suppliers can provide. This circles back to supplier diversification and, to degrees, raises the question of whether Vodafone’s move with Nokia signals a successful pressure campaign to keep doing things the same way, but cheaper.
There’s also an optimistic view, shared by a TIP higher-up (who also acknowledged the cynical position) that Nokia working on Open RAN does indeed represent supplier diversification in that it’s a supplier that hasn’t previously sold Open RAN kit. Another industry watcher, referencing mandates to rip-and-replace gear from Chinese vendors, wondered on the sidelines of the show if TIP had created a “virtual Huawei.” Regardless of interpretation, there was a bit of conFYUZion today.
