How Robin.io is Making Waves in Open RAN, 5G
Partha Seetala, CEO at cloud native networking specialist Robin.io, sits down with Ray Le Maistre of Light Reading to discuss the key architectural changes underway in telecoms networks as operators implement their 5G strategies. Partha also discusses how the deployment of the Robin.io platform as part of the Open RAN-based 5G rollout at Japan’s Rakuten Mobile has enhanced the company’s credibility in the global telecoms market.
[Ray Le Maistre, TelecomTV] The deployment of more Cloud-oriented communications networks is attracting new names into the telecom sector. So, I’m talking today with Partha Seetala, he’s the CEO at Robin. So Partha, Robin is a fairly new entrant to the telecom space having initially made its name in the enterprise sector. How did Robin find its way into some of the telecom industries’ most cutting edge networks?
[Partha Seetala, CEO, Robin.io] First of all, thanks Ray for having me on your show. You’re right, we have been a pioneer in the Cloud-native space on the enterprise side. That’s where we started and we have several large financial institutes that have been using Robin to accelerate their Cloud-native journey for several of their complicated network storage-intensive applications. So, the interesting thing with telecoms for the most part have been using legacy type of infrastructures for a very long time. The first wave of virtualization also was done on things like OpenStack complicate a very long project and so on. Now, with our background in the enterprise space, bringing these complex workloads to Cloud-native, what’s happened is that, I mean, we found an interest from some of the pioneers in the telecom sector who have been looking at transforming how they will be doing rolling out 5G as well as edge applications using Cloud-native technologies. And the work that we have done in the enterprise when they looked at that, they basically said that the platform that Robin has built is very powerful. And that is something that could be used. They actually actually accelerate the roll-out of 5G services as well as edge services using a more modern technology like Kubernetes and automation and things like that. So, it started off with the pioneer in the space, Rakuten Mobile. When they were looking at launching their 5G services, they did a bake off, they looked at different solutions and that’s how they picked us. And that’s how we essentially entered the 5G space, or the Cloud-native or telecom space.
[RLM] So that deployment to Rakuten Mobile in Japan has certainly helped to put Robin on the map. How important has that been in terms of gaining credibility with other telecom operators?
[PS] So the deployment of Rakuten has been very, very important for us. Clearly, Rakuten is a pioneer in this space. They are the first to envision 100% Cloud-native architecture for deploying their 5G and edge services. And of course, Rakuten has brought a lot ecosystem partners to get this services rolled out. Being part of the Rakuten ecosystem set of partners, we were able to go and work with several of the leading vendors in the space. And again, given that Rakuten has been very religious about sticking to the principles of Cloud-native and instead of mixing Cloud-native with legacy architectures. Because of there requirement that we stick to 100% Cloud-native architectures, we have been able to harden the platform a lot. And now that, of course it’s been running in production in their network, it has definitely helped build the credibility of Robin significantly. Not just among the other operators but also among the ecosystem because they can actually see that the Robin Cloud-native platform is a buildup of scale, is a buildup of resilient and it meets the SLES that are required in a large scale network, such as Rakuten’s.
[RLM] Key to such platforms, automation and orchestration. And these have been hot terms, buzzwords in the industry for a while now, but why are automation and orchestration so important for the success of 5G?
[PS] So again, 5G, again, it’s not an incremental improvement or 4G we’re talking about. A lot more end points or UES, or in fact, even sensor type of devices connecting all the network. So, of course you need to have a lot more bandwidth then that bandwidth also has to be delivered at lower latency and so on. Which means that there’s going to be a huge proliferation of cell sites, edge locations, and so on. So it’s going to be an order of magnitude higher than what we have seen so far in the 4G space. But when you have such a huge volume automation becomes key. One of the obvious reasons for automation is how quickly can you go and roll out a new cell site? How quickly can you go and deploy a new edge servers closer to the end point? Now, doing this manually is going to be very difficult, especially when it comes to that scale. So, that’s definitely requires a level of automation that the telecom industry has not really seen until recently. And of course, without doing that, you’re essentially increasing your operational expenses significantly. But there are also other aspects of automation that become important, which is around, you have your services up and running and you’re running low on, let’s say capacity or bandwidth, and you would like to automatically go and scale. And what you really want to do there is you wannna set up policies that are managing, or that are essentially monitoring the usage and based on the usage, you have to trigger certain actions, which could be things like increase the compute capacity or increase the network capacity at a particular location. Now ideally, you want to just set it as a policy and forget about it so that even in the middle of the night, there’s a spike or there is something that requires additional capacity it’s automatically taken care of. And of course, when the load goes down, you want to take it away, and that’s through Cloud-native by the way. Now without automation, how do you even go about doing something like this? You would essentially have to have humans that are constantly monitoring your network and constantly adjusting it? That’s the other aspect of automation. And I would say the third aspect of automation is around doing things like faster service roll-out, which I already explained. But also things like how do you go and do upgrades of these applications? As in when your network service function vendor is giving you newer bits. Because you want to be at the cutting edge of all these applications. But when you’re talking about hundreds and thousands of servers, tens and thousands of cell sites. I mean, there is no way that somebody could manage this without a full-blown automation. Where you just basically, again, define certain policies and have these automation thing, well, go roll-out. All these services, upgrade all these services and stuff. So, automation is critical. Without automation it’s going to be A expensive and B also it’s impossible to ensure correctness when you’re doing these roll-outs.
[RLM] So Partha, in a short amount of time Robin has built up a long list of partners. Can you explain the importance of having a robust ecosystem of technology partners?
[PS] So, we have been doing Cloud-native for several years now and right from the get-go, our focus was to ensure that we can run these complex data storage-centric workloads, and network-centric workloads in a Cloud-native manner. I mean, unlike some of our competitors who are focused on just hardening Kubernetes and offering it as a solution to the enterprises, we actually took a very different approach that to actually succeed at Cloud-native, you need to have the ability in the platform to go and run these complex workloads that are coming in from the application vendors across various spectrum of use cases. Now, as part of this, I mean, we have been building this platform. We have built a lot of technology, a lot of IP, we have got 65 plus patents in this area of storage, automation, network management, scheduling and so on. Now, that gave us a lot of advantage because of the technology underpinnings in our architecture, you are able to go and onboard a lot of partners. As I said, across the various spectrum of use cases. Now, this is actually beneficial because in the Cloud-native space you really need to have of course, a platform that can support all these things, but also the application vendors. We’ll also need to have a platform that has these capabilities. And it’s the journey on the 5G and the edge site. It started with Rakuten, off course Rakuten had a lot of partners that it was working with. And because our platform was being hardened over the last few years to support these complex workloads, it became relatively easy for us to go and onboard on Windows, the Core Windows, the mobile edge application vendors, the OSS vendors and so on. The innovations that we have been putting into the platform had started to pay dividends at Rakuten. Now, what’s happened is that because of the broad spectrum of workloads that we could onboard, and of course fine-tune the platform while we’re onboarding this at Rakuten, it has led to other partners also to collaborate with us. We have of course, partners across the spectrum from the hardware vendors to the chip manufacturers, to the PGA vendors, to the LAN vendors, to the Packet Core vendors, to the OSS, BSS vendors, to the edge application vendors and all of those folks who have essentially now certified on the Robin platform. And now, this essentially helps everyone. It helps the operators because they can now very quickly go and onboard one of our servers that they want onboard very quickly because the integration effort is less. It also helps the partners because now they have a robust Cloud-native platform that has been tested at scale in production network. And for them, it’s very easy to go into an operator and then roll-out there services. So, I think it’s a win-win across the board and that it’s been a very, very, we’ve been very fortunate that we have been able to validate this at Rakuten Mobile. But we are really fortunate also that we have had now the ability to bring in so many different partners to run on top of the Robin Cloud-native platform.
[RLM] So, lots of exciting developments at the moment. What can we expect to see from Robin in say the next six to 12 months?
[PS] So we have always been innovating. And right now we are years ahead of where the competition is when it comes to Cloud-native platforms. We are the first, and as far as we know, only Cloud-native platform that has a live production grade 5G deployment on it. And Rakuten has been a pioneer, it has been pushing us a lot in terms of ensuring that we are actually compliant in the Cloud-native way. It’s not like we are compromising it. And Rakuten has been very religious in that matter that they wanted this to be pure play Cloud-native implementation. So again, the first and only platform that has been able to support this in production, and this is of course, given us a big lead over any other platform out there. But again, we continue to innovate. We are working very actively on things like network slicing. And I already talked about the innovations that we have done, the automation area. We are extending that to a much, much more complicated, more AI driven automation so that even the policy scan, can be essentially automatically discover the LAN and then apply it. We are working on things like how can we further decrease the OPEX and CAPEX for an operator so that it becomes more economically viable to go and run these massive scale 5G networks in production? So, a lot of innovations that we’re working on, a lot of interesting things that you can expect to see from the Robin team in the next few quarters.
[RLM] So, that sounds like a very exciting 2021 ahead for Robin. Partha, thanks very much for sharing the company story with us today. Thanks very much.
[PS] Thanks Ray, it has been a pleasure talking to you today.