Earlier this year, Red Hat, Intel and Ericsson announced a collaboration to drive a 20% power reduction in the cloud RAN digital unit while achieving the same performance. This is turning technology into partnership and into real-world impact in operator infrastructure, accelerating enterprise digital transformation across industry verticals.
Honore La Boudette, Vice President of the global partner ecosystem success team at Red Hat, and Jenny Perrovian, Vice President and General Manager of network and edge compute at Intel Corporation, join Fierce for an engaging discussion about this partnership.
The conversation explores three topics within cloud infrastructure - automation and AI in the cloud, cloud sustainability, and cloud for vertical industry applications.
Cross-industry collaboration and solution development are forcing changes in the way that solutions are delivered due to complexity in the supply chain and coordination of software and hardware solutions. Collaboration between Red Hat and Intel enables foundational platforms that adhere to cloud principles while delivering industry-specific cloud solutions that sit on top of that foundation.
Listen to this exclusive episode, read the transcript below.
Steve Saunders:
Hi, I am Steve Saunders on Cloud 9. 2023 has turned out to be a huge year for cloud infrastructure, as the technology matures and continues to alter the way companies do business, increasing efficiency in productivity, and reducing CapEx and OpEx. But this is still a hugely complicated and challenging market. Now as the founder of Silverlinings, I get to see that play out firsthand in the traffic analytics on our site, which show that there are three topics which cloud network architects are really fixated on currently. And they are automation and AI in the cloud, cloud and sustainability, and cloud for vertical industry applications.
So today on Cloud 9, we're going to drill into these topics with the help of experts from two of the leading companies in cloud. Let's meet them now. Honore LaBourdette, VP Global Partner Ecosystem Success Team at Red Hat. Hi, Honore.
Honore LaBourdette:
Hi.
Steve Saunders:
And Jeni Borovian, Vice President and General Manager Network & Edge Compute at Intel Corporation. Hi, Jeni.
Jeni Barovian:
Hi, Steve. Great to be here.
Steve Saunders:
Honore, is sustainability an area where Intel and Red Hat have collaborated? And how do you measure success in that critical area?
Honore LaBourdette:
Thank you, Steve. Sustainability is a key topic inside of the telecommunications industry today. Currently, the industry uses two to 3% of the total power consumption of humanity. Energy costs can range from 20% to 40% of OpEx costs for service providers. And with a four X increase in mobile data traffic by 2027 and multiple fold of 5G radio footprint, we expect the cost of energy is going to become a major challenge for telcos. Hence, making energy optimization crucial for every layer in building blocks of the network.Intel and Red Hat are prioritizing building RAN and core ecosystems by utilizing capabilities in Red Hat software and Intel hardware to enable telco service providers to develop and optimize energy efficient networks. And we're extending it out to all of those suppliers to the telecommunications industry to participate in our work.
Steve Saunders:
Great. Thank you. Jeni, anything to add on this one?
Jeni Barovian:
Yeah, absolutely. This is a huge area of partnership between Intel and Red Hat, and really across the value chain. If we look at the responsibility that the ICT industry has, it's really a responsibility to address our own carbon footprint goals, which we call TechZero, and also an opportunity to enable emissions reduction in other industries. So, we look at this as tech positive.So within Intel specifically, we've had a longstanding commitment to address our own footprint, our own TechZero. We've made a commitment to achieve net-zero greenhouse gas emissions in our own operations in scope one and two, by 2040. But when we look at scope three, that's really the bigger challenge. This includes our supply chain as well as downstream value chain by really driving increased energy efficiency, and lowering the total carbon footprint of our products and platforms, which in turn, help our customers achieve their sustainability goals.And like Honore said, two of our key areas of focus are in the RAN and the core. So as we look at how to measure success, we really have to look at those deployments of energy optimized solutions across networks and across Edge computing deployments, and see if they're really driving to those goals of lower emissions and total carbon footprint. So if we look at what we've achieved so far in the cloud RAN, earlier this year at Red Hat Summit, Intel and Red Hat and Ericsson announced a collaboration to drive a 20% power reduction in the cloud ran digital unit while achieving the same performance. So, no compromise in performance while achieving power reduction. If we look at what we've achieved so far in 5G Core, similarly when we look at leveraging the 4th Gen Intel Xeon scalable processors, we've been able to enable operators to reduce their energy costs and carbon emissions by delivering an overall power savings of 30% on the CPU.And so really what this is, it's that collaboration between hardware and software. The power of software from Red Hat and other ISVs and other players across the ecosystem in order to access low level hardware features like our P-states and C-states on our processors to control power consumption in real time, based on what's happening on the platform while still achieving the KPIs that operators need to achieve. And still maintaining performance and service level. So, it's really about turning that technology into partnership into real world impact in operator infrastructure.
Steve Saunders:
Great. Thank you, Jeni. Honore, I'll come back to you. Is one of the areas of collaboration that you have with Intel to deliver cloud solutions within vertical industries? At Silverlinings, we're obviously writing about that and hearing a lot about it from our audience. And I think some of them are wondering whether there are changes in the way solutions are delivered for different vertical markets. Do you have any examples of that, and which industries are the first movers here?
Honore LaBourdette:
We do. Red Hat and Intel have been collaborating to accelerate enterprise digital transformation across industry verticals for some time now. For example, we have co-developed a scalable, flexible, and high performance platform for enterprise, hybrid, and multi-cloud workloads. With this reference architecture, enterprises can move to a modern cloud native infrastructure meeting today's IT demands.Jointly, we've taken the best practices from the telco industry and have been prioritizing manufacturing, retail, and energy verticals. We have built software-defined Edge architecture supporting this transition. We're also working with our customers as well as the ecosystem to increase awareness of our joint industrial Edge solutions utilizing AI and computer vision technologies.
Steve Saunders:
Great. Jeni, what's your take on vertical industries in the cloud?
Jeni Barovian:
I think Honore said it really well. Our collaboration between Intel and Red Hat has enabled foundational platforms that enable the cloud principles to be employed to these vertical industries, as well as specific cloud solutions to be developed that sit on top of that foundation. And so, I'm going to pause for a moment. We started talking about communications a little bit. I want to talk a little bit about a collaboration that we had with the cloud native cable modem termination solutions, so virtual CMTS. And that utilizes Red Hat OpenShift and it utilizes Intel's Xeon processors. And also leverages other technologies from Intel like Quick Assist Technology and our Ethernet 800 Series.And what this did was it helped multi-system operators or cable operators to decouple workloads from specific hardware by virtualizing those network services and encapsulating them in containerized network functions. And we really developed this into a solution blueprint, which provides these operators with the platforms that can respond to today's needs while being ready for future needs. So, it was that collaboration that we built into a blueprint that's enabling other operators to scale.If we dive a little bit into some of the vertical industry impacts as well, we collaborated on a cloud native industry 4.0 industrial control system solution by combining Red Hat OCP with our Intel Edge Control Software for industrial, as well as Intel Xeon and Intel Core processors to decouple that control system hardware and software to provide a gateway to open and flexible ecosystem solutions. So really what this is, are just a couple of quick examples of the types of collaboration and solution development that's happening across many different domains that's really embracing the fact that this transition to cloud-based solutions is forcing a change in the way that solutions are delivered.The supply chain is so much more open and so much more complex. And it really takes a village in the ecosystem to coordinate across the hardware and the software and the services and support to align all of those pieces to meet the customer needs. And really in our partnership between Red Hat and Intel, we're at the heart of driving this alignment across the value chain. Whether it's equipment manufacturers or software vendors, developer system integrators and so on, so that open server-based cloud ready solutions are available and easy to use across the vertical industries.
Steve Saunders:
Okay, great. Thank you. Let's switch to another really hot topic at the moment, which has been the subject of a lot of mainstream media coverage as well as technology press coverage, which is AI. Jeni, I'm going to come to you first on this one. What are your views on the advent of AI enabled network automation? Is it a good thing? And is this in your wheelhouse, as partners with Red Hat?
Jeni Barovian:
Yeah, this is a huge growth area in AI enabled network automation with use cases that are ranging from local closed loop automation at an individual node level through network-wide automation and optimization. I kind of go back to what I talked about in the sustainability part of our conversation. We had an opportunity to kind of real-time monitor what is going on on the platform performance KPIs and then apply power management technologies in order to lower that power while maintaining performance. If we take a similar type of approach and apply AI technology, we have an opportunity to really accelerate the impact.So we've got a great example that we demonstrated at Mobile World Congress Shanghai, that most recent show, where we had a centralized RIC. And we applied AI to the RIC to control the P-states of multiple RAN nodes to achieve energy savings. And so, it's a little bit different from the collaboration I talked earlier, because this is about AI at the central controller to take in that data. But then really turn it into a predictive capability to monitor activity, and then ultimately predict activity and then minimize power over time. So, that's a great example of taking that foundation and infusing AI to increase the impact.Another great example is if you look at the need to drive traffic analysis and deep packet inspection in the 5G core, at Intel, we've released a traffic analysis development kit, we call it TADK, to add AI prediction to 5G UPF DPI engines. I think as everyone here, everyone listening probably knows DPI is a key capability to monitor network traffic in core networks. And increasingly, that traffic is encrypted. And so, AI machine learning enables that traffic analysis development capability to inspect that encrypted traffic in flight without impacting performance. So, it really monitors that traffic through examining those encrypted packets that are coming in. And then it learns those topological characteristics and builds models to classify it, which then in turn has the opportunity to apply that traffic analysis capability to broader network automation across the infrastructure. So, a lot of really great examples of foundational technologies. And I really look forward to Intel and Red Hat collaborating on turning these into commercial solutions.
Steve Saunders:
And those aren't examples really of taking jobs away from human beings. They're just examples of automating things which just don't exist. Right ---
Jeni Barovian:
Which we need to do better. Exactly, because ---
Steve Saunders:
We need to do better with a real network and business outcome. Honore, what do you think about the AI market at the moment? Well,
Honore LaBourdette:
I think Jeni made some excellent points on this topic. Network and process automation help CSPs reduce the cost of their operations, and moves compute closer to the edge where the data is actually produced. Now AI supports telco service providers capturing that data from the network interactions of thousands of network touchpoints, in particular in the RAN space and with our customers.We believe the future of AI is open, hybrid, and will extend to the edge. And as Jeni had pointed out, red Hat and Intel are working together to provide enterprise AI, ML solutions to our joint customers with a developer program. Having a trust foundation reduces the friction throughout the lifecycle. Red Hat Ansible with its automated event driven ability is able to connect network intelligence analytics service requests from IT solutions to take an automated action.Ansible Lightspeed makes specific automation knowledge and skills more broadly accessible, and domain specific AI singularly focused on automation. Ansible helps the service providers to rely on their network data to take faster action, address corrective action, thus reducing their cost of OpEx. And we will continue to work with Intel on a number of projects where we can enhance AI and ML to better improve both the OpEx, the CapEx, and the availability of the capabilities that service providers are wanting and required to deliver to their customers.
Steve Saunders:
Great. Honore, let me stay with you just to talk about another component of cloud, which is cloud native, which has really enabled a new way of software development and business opportunity. And it's brought together in this unique way, a growing ecosystem. You obviously work very closely together, but are there criteria for admission to the Intel Red Hat Ecosystem Club? What are the criteria for being part of your ecosystem? And can you point to some of the benefits for your customers?
Honore LaBourdette:
So together, we take the complexity out of cloud native. This enables faster collaboration and innovation. Working closely with Intel, we integrate software and Intel's hardware stack closely to maximize the value of hardware and software for other applications that run on top. Integration and certification of this stack takes the risk out of deployments and operations. Together, we have our CNF Certification Program to reduce deployment time and complexity. Red Hat and Intel created a cloud-based onboarding service in a lab test bed supporting both CNFs and VNF environments. This level of integration increases security, reliability, and improves the ongoing support in day two operations.The importance of Intel and Red Hat to work closely with ISV partners in advance of production deployment is crucial to the success of telcos. As we are transitioning to cloud native containerized networks, we need to continue to build an ecosystem that is inclusive of all involved parties, Silicon, Hardware, OSV, network workloads provided by the ISVs, and ultimately SI partners pulling this altogether as a complete deployable cloud native solution.
Steve Saunders:
Jeni, what do you think? What's the criteria for admission to the ecosystem?
Jeni Barovian:
The promise is huge, of course, right? So if we look at the promise of cloud native deployments, it brings network and Edge customers, the ability to drive greater flexibility, cloud scale efficiency, new business opportunities through optimized software, and a more modular design to their systems. Running on Intel's cloud native optimized hardware and Red Hat's cloud native optimized software, I talked about a couple examples before, virtual CMTS, virtualized containerized industrial solutions. Another great example is what's happening in the vRAN, which is really a distributed design with functions running in containers on Kubernetes based infrastructure, leveraging underlying hardware capabilities, specialized acceleration. Intel's vRAN Boost is just one of many examples.And so, to leverage that specialized hardware capability, it really takes a village in order to drive the software capabilities and create a system approach to drive the outcomes that customers are looking for. And so like Honore said, it's about taking the complexity out of this transition by together driving the ecosystem towards coordinated delivery of the solutions. It is definitely a group effort, a community effort that's required. And so the cost of admission, Steve, I guess is what you're asking, is definitely a willingness to participate in this very complex ecosystem in order to drive better outcomes and all of those promises of flexibility and efficiency for our customers. And so, we certainly welcome collaboration from everyone across the value chain in order to drive these great results.
Steve Saunders:
I like the fact that you're not pretending it's easy, and there's no cloud washing going on on the show today. You guys are seizing the -- of cloud complexity and turning it into something which is really valuable for the communications industry. And I tip my red hat to you for doing that.This is really interesting. I think our audience, which has been bombarding us with questions about automation and AI and sustainability and vertical industry, I think they will have really got a lot out of this. So Jeni from Intel and Honoree from Red Hat, really want to appreciate you. I really want to thank you for being on the show today. We appreciate you. Thank you.
Honore LaBourdette:
Thank you, Steve.
Jeni Barovian:
Thanks, Steve.