Are communications service providers on the cutting edge of the rapid and aggressive swing toward cloud-native networks? CSPs have been migrating and managing business-critical, and even mission-critical, applications in the cloud to help run their business. But CSPs are just beginning to leverage cloud into their most prized asset: the network.
Adjusting the architecture and deployment of telecom networks is no small task. However, technology advancements are necessary to help CSPs keep up with the ever-increasing demand for bandwidth. To fully leverage and optimize the benefits of a cloud-native mindset, a new philosophy—and a new solutions architecture—is necessary. Specifically, CSPs should adopt a cloud-native network automation (CNNA) architecture spanning the full spectrum from core network functions to radio access networks running on cloud and cloud edge.
Why Cloud-Native Network Automation?
Network automation is not a new concept. Management and orchestration (MANO) were central themes of network functions virtualization architectures. That said, full implementation of MANO has usually been an afterthought, and the value never fully materialized. Complex alignment among vendors, services, network functions and infrastructure was just one of the issues. The introduction of 5G, Open vRAN, Multi-access Edge Compute and highly distributed cloud-native architectures are putting immense pressure on network operations to scale while controlling costs.
Automation is now returning to the forefront. But as the network transitioned to cloud-native, many have tried to port MANO concepts instead of adapting and leveraging the full potential of cloud-native best practices and technologies. It’s time to fully embed automation into the network fabric, from software to infrastructure, across domains and vendors.
Developing Standards Through Nephio
CSPs would like to align with cloud-native practices and technologies, but there are still fundamental differences between workloads for telecommunications and enterprises. CSPs’ network functions are more technically complex; they handle millions of subscriber sessions or heavy data throughput, requiring specific infrastructure and networking optimization. The operating model will also differ based on the workload criticality and the number of infrastructures, clusters and sites to support.
In that context, Google Cloud is a charter member of the Linux Foundation’s open source project, Nephio. The focus of Nephio is to reduce these frictions and accelerate network cloud-native automation adoption. It aims to deliver carrier-grade, simple, open, Kubernetes-based, cloud-native intent automation.
- It uses standard automation templates that materially simplify the deployment and management of multi-vendor cloud infrastructure and network functions across large-scale edge deployments.
- Nephio enables faster onboarding of network functions to production, including provisioning of underlying cloud infrastructure with an authentic cloud-native approach.
- It also reduces the adoption costs of cloud and network infrastructure.
Achieving Cloud-Native Network Automation in Telecommunications Together
This paper explains how project Nephio, in partnership with leaders across the telecommunications industry, seeks to achieve true cloud native automation that will help drive scale, efficiency, and high reliability across network operations.
Download NowExpressing an intent means that the network function focuses on stating its resources and configuration needs, not all the underlying commands. For example, the network function needs X capacity at Y location when Z event occurs. Kubernetes, through controllers and actuators, will continually ensure that the intent is served. Since the same declarative model applies to network function resources management, its configuration and the cloud infrastructure, Nephio removes misalignment issues and the entire stack can be automated using the same approach.
Another essential element that Nephio offers is the standardization of Kubernetes templates (operators/custom resource definition) to simplify multi-vendor interoperability between each layer of the infrastructure.
Dozens of leading industry players—from cloud service providers and CSPs to chip manufacturers and network infrastructure suppliers—are participating in Nephio’s work to develop and promote a vendor-neutral standard for CNNA.
Google Cloud and Intel: A Powerful Platform for CNNA
To properly implement and reap the benefits of a CNNA architecture, organizations need a common cloud-to-edge platform, powered by cutting-edge technology. Google Cloud and Intel have long collaborated to build modernized, future-proof solutions for a wide range of use cases.
CSPs can take advantage of high-performance networking—including turbocharged 5G network infrastructure—using Google Distributed Cloud Edge powered by 3rd Gen Intel® Xeon® Processors. This helps service providers engineer and launch new services in a highly automated, intelligent framework that substantially lessens the management burden.
Solutions from Google Cloud and Intel promote increased visibility, automated management and high performance at scale for a truly carrier-grade networking experience. Google Cloud and Intel have fully embraced the principles and work of the emerging Nephio standard for CNNA. In fact, both companies are founding participants in the Nephio initiative.
Conclusion
Cloud-native represents both the present and the future of network functions and infrastructure. CSPs’ network infrastructure must be fully aligned with all cloud-native philosophies in a highly intelligent, automated, performant, secure, resilient and scalable solution.
Google Cloud and Intel provide GDC Edge as a fully managed solution to run network functions and applications. GDC Edge is purpose built for the cloud-native paradigm, while providing a clear path to the future with a commitment to the Nephio emerging standard for CNNA.
