Did Open RAN deliver on the technological and market-based promises it was meant to bring to? Tough question. And what were those initial promises again? We went back through the O-RAN Alliance archives to a white paper published in October 2018. At that time, the group established to lead the way forward focused on a two-pronged vision—openness and intelligence.
The openness piece was about “building a more cost-effective, agile RAN;” open interfaces would “enable smaller vendors and operators to quickly introduce their own services…[and] customize the network to suit their own unique needs.” This multi-vendor kit would enable “a more competitive and vibrant supplier ecosystem” capable of delivering “more democratic and permission-less innovation.”
As for intelligence, the alliance identified automation as the way to “tame..complexity;…networks must be self-driving, they should be able to leverage new learning-based technologies to automate operational network functions and reduce opex.” The idea was to “embed intelligence in every layer of the RAN architecture.” Using those open interfaces, “AI-optimized closed-loop automation is achievable and will enable a new era for network operations.”
So how’d all that go? We asked a panel of experts who, generally, viewed Open RAN as a work in progress marked by significant achievements but with the best yet to come. Also of note, disaggregating the RAN is part of larger network transformations involving everything from cloud-native core and fiber densification to SMO evolution and BSS/OSS modernization. And, of course, AI has a huge role to play across networking domains.
Over to Wind River’s Randy Cox, vice president of product management. Reflecting on those wide-ranging initial promises, he distilled it to flexibility, innovation and total cost of ownership (TCO). “On flexibility, if you look at that, I think we’re experiencing that in carriers who have actually decided to…deploy Open RAN.” Armed with the ability to mix-and-match best-of-breed hardware and software gives “the carrier the power to decide the make up of their network.” He continued: “In terms of innovation, I definitely believe it’s created because you have different vendors contributing and bringing a wealth of creativity and technology together that starts to feed on each other.” As for TCO, Cox acknowledged the “competitive aspect” of that, but added that “new technology, as it advances, that also brings in the element of reducing total cost of ownership.”
Mavenir has been a driving force in Open RAN since before we called it Open RAN, and the company has pushed hard to achieve broad interoperability with other vendors. The company’s Rick Mostaert, vice president of product management, said vendor interoperability has been “achieved times 10…The ecosystem is huge…We see not just open fronthaul being implemented but multiple cloud stacks, multiple Layer 1 protocols, multiple servers, etc…It’s up and down the entire technology stack where we’ve done integrations with other vendors. And that’s just great for everybody. It allows new people to come into the industry which is desperately needed.”
As for the market dynamics, Mostaert said that bit is “more complicated. Not everybody is bought in…There’s still an element of vendor lock-in.” That said, he described the issue as “not a technology issue. It’s a will issue.” Long term, however, “Operators are just not going to accept the status quo…and that market dynamics piece of it will evolve…I see great things ahead but it’s been probably a little bit slower than we would’ve liked.”
There’s a relatively new area of discourse within Open RAN around the idea of “single vendor Open RAN” wherein the vast majority of the radio system is provided by large incumbents with limited integrations of third-party radios. To pick that apart however, these same implementations, in addition to featuring third-party radios, also feature variety in servers and chipsets. That to say, you could make the case that single-vendor Open RAN isn’t necessarily an accurate descriptor, but it is also not a full-blown multi-vendor implementation. So what does that suggest for the initial vision of multi-vendor?
Cox said, “There are stages and a progression in this journey.” Moving to Open RAN is a big, long-term decision operators have to make, he said. And “once they make that decision, they have to actually execute and move forward in a way that is fast-paced but also achieving the goals that they want. There’s a basic network level that you need to be able to achieve obviously.” From there, Cox said, the focus is on time-to-value and delivering new innovations enabled by an open architecture. “I do believe that bringing in new technology will be a lot faster paced…You do have to get over the initial hurdle, but after that definitely the time-to-market in terms of bringing that new capability is definitely accelerated.”
Single-vendor Open RAN isn’t a phrase in his vocabulary, Mostaert said. “The ecosystem that is being built up around Open RAN is actually one of the most exciting things about it…It’s just a force multiplier for everything. Now is everything perfect where some small company can just come in and just join and provide some product? Not yet. But the architecture is there, the platforms are there, the open architectures are there, and it’s all been proven. Now it’s just business.”
And beyond delivering benefits to operators, Mostaert said it’s also “an absolute advantage for a company like Mavenir” who can choose their own preferred partners at different points in the technology stack. “We have ultimate choice…There’re different ways to deploy, the ecosystem is growing, it’s very powerful, and the amount of money, of capital, and engineering talent that is now in the wireless network infrastructure business through Open RAN is a great thing for everybody—consumers, operators, vendors. I just see it getting better and better.”
From open interfaces to open-ended innovation
So what’s next for Open RAN? There’s certainly global investment and support, a strong standardization apparatus, and capabilities that are consistent with the early vision for 6G.
“I think that as we look forward, I do think that, in the terms of 6G…that many of the topics we’re discussing here in 5G will in some sense be table stakes for 6G,” Cox said. “New technologies like cloud-native networks, automation, AI energy and efficiency, those are going to be absolutely required in 6G. He also pointed out that the deployment of new spectrum with 6G plays to the software-based flexibilities built into Open RAN, particularly around a unified infrastructure capable of supporting multiple generations of cellular. And, perhaps most importantly, Cox called out that successful technology transformations come in tandem with successful workforce transformations. Carriers are “going to be needing to plan and really build their resources to handle a new network like this. That requires upskilling their teams, broadening their capabilities and really retooling their workforce for new areas of advanced technology.”
Mostaert divvied the outlook into three parts: interoperability, scale and innovation. As far as proving out open interfaces, “We’re well pas that,” he said. On scale, “We’re seeing larger scale deployments…That’ll continue through this year and into next year.” On innovation, “Now that we have the basics of the technology, we have the vendors that can scale Open RAN, operators that know how to run it…we’ll see the power of the ecosystem, of the innovation…We’ve come a long way in five years. The basics are there. Now it’s time to innovate and see where we can take this.”
