Three principles for bringing hyperscale cloud models to service provider networks (Reader Forum)

“The cloud eats away at the world.” At least that’s the big conclusion from the technology industry these days. And that’s a good thing. For users, the cloud means having access to digital services whenever and wherever you choose, without worrying about how everything is happening behind the scenes. But what if you are responsible for everything happening behind the scenes? If you’re part of a service provider organization, what does cloudiness actually look like?

Overall, you can expect data center (DC) architectures to become the main template for a growing portion of your portfolio. The good news is that cloud DC approaches are some of the most mature and proven in the industry, initiated by hyper-scale leaders such as Google, Microsoft and Amazon. Now these powerful cloud concepts can play a much bigger role in your business.

What does the growing reliance on cloud DC mean for service providers’ operational teams? And what should you think about as you move through this transition? Let’s take a closer look.

Understanding the service provider’s data centers

Data centers have played a role in both internal and customer service providers’ operations for years. The difference is that now most of the subscriber network itself – from core to end to radio access network (RAN) – is virtualized. As more and more of your network becomes software running commercially available servers (COTS), cloud DC concepts are becoming applicable to more and more of what you do.

But DCs are not monolithic. Today, service providers use DC in three main ways:

  • IT data center: Functionally, service providers are large enterprises and like any business, they must support various enterprise applications such as Salesforce, Microsoft 365 and many others. However, DC service providers also support applications that directly affect subscribers, such as operational support systems and business support systems (OSS / BSS). So, even as more applications move to the public cloud, IT DC infrastructure will continue to play an important role in business.
  • Managed Services Data Center: Service providers use these DCs to provide managed services to their (usually corporate) customers. This can include things like managed web security, software-defined wide area network (SD-WAN), secure access service (SASE) products, and more.
  • Cloud / Edge Cloud Data Center: This is the area where the biggest change is taking place, as operators are adopting Telco Cloud architectures to support basic telecommunications loads. In core networks, this includes virtualization and cloud features such as Developed Packet Core (EPC) and IP Multimedia Subsystem (IMS). On the edge, operators are clouding RAN sites to enable Open RAN concepts, as well as new enterprise services based on finite calculations, such as content delivery networks, affiliated factories and more.

We are not yet at the point where operators converge these different DCs into one fabric, and we may never get there. These DCs can perform extremely different types of loads with very different requirements, requiring a variety of DC topologies and presenting CSPs with imposing operational challenges. Although design can and should vary, common DC fabric management and automation systems, along with software-defined network tools, can deliver the operational consistency that provides the business flexibility that CSPs strive for. This means you can now take a page of hyperscalers and replace yesterday’s slow, inefficient operations with modern cloud approaches.

Key principles for scaling your data center’s footprint

Cloud approaches promise great benefits. When you can start your network as a pool of flexible cloud resources, you can deploy new services much faster. You can provide better performance and reliability to customers and you can manage your network more efficiently, at a lower cost. At least modern cloud approaches I can provide these benefits. But just because you manage more of your network like DC, doesn’t mean you’ll automatically realize them.

When planning your cloud investment for the coming years, keep the following principles in mind:

  • Build with thinking in the first place for experience: While providing a more demanding mix of services, the cloud allows you to more easily move capacity where it is needed. But capacity only takes you this far. Especially for telecommunications / end cloud operations, it’s less about how many bits you can push per second and more about the experience you provide. This means not only the experience for your end users, but also the experience that your DC operations teams face. The life cycle management on day 0/1/2 cannot be delayed. But using infrastructure tools like code to speed up DC planning, design, and configuration is just the beginning. Real savings come with simplifying current operations. Ensuring that operational teams can consistently and reliably manage the infrastructure and provide a great customer experience, whether internal or external, should be the main focus of any design decision you make.
  • Intentional-based automation is now essential: Service providers’ networks are becoming more complex, and the new 5G and border services are making them exponentially larger. The only way to tame this complexity is with automation. But too many automation approaches focus on accelerating outdated manual processes instead of replacing them. Yes, you can use scripts to speed up CLI configurations, but people still need to understand the step-by-step techniques for these scripts. To run your business more like a hyperscaler, you need to completely abandon this model. Instead, you need intent-based automation – automation tools that use AI engines and machine learning to receive vast amounts of real-time information for your network. The operational way of thinking must shift from the work that needs to be done to the decision that needs to be made. Telemetry based on graphical databases, which are characterized by the correlation of vast amounts of information, help this transition. In addition, state-based approaches, rather than event-based systems, also help sharpen decision-making.
  • Give priority to the operating sequence: One of the biggest advantages of 5G is its extension to open standards and APIs, which allows service providers to represent more providers in more parts of the network. This is great news – more suppliers means more competition, more innovation and a better economy. But if you’re not careful, it can also mean huge operational complexity when different parts of the network require different vendor-specific tools and processes. Make sure that no matter how many providers you introduce in your environment, you can use the same sequential operations. Your DC automation tools should be able to manage infrastructure from any provider and even a white box with ease. Operational consistency is a cornerstone of the hyperscaler strategy and the key to their success. Consistency leads to reliability. Reliability leads to speed. And not just “speed” in the narrow sense of process automation, but in the broader sense of speed of operations.

Huge, diverse service provider environments will never be exactly like homogeneous hyper-scale clouds. Fortunately, there are now open, intent-oriented automation solutions designed specifically for service providers. They allow you to take the most innovative hyper-scale operational concepts and apply them to the unique requirements of telecommunications DC networks. As long as you stay focused on what matters when you create next-generation DC, you can bring the cloud revolution to your business.