Skip to main content
Robot sitting with data center switches

Data Center Fabric

Make data center switching and cloud environments easier to scale, adapt and operate

What is a data center fabric?

A data center fabric is a network of leaf and spine switches that work together to provide a resilient and scalable infrastructure for connecting software entities installed on servers. Each leaf and spine switch includes hardware and software. The hardware provides the physical networking interfaces, switching matrix, control complex, fans and other physical components. Each switch runs a network operating system (NOS) that provides the intelligence, routing capabilities, telemetry and other programmable capabilities needed to operate the system. The network is typically managed by data center fabric management and automation tools or platforms.

What is the Nokia Data Center Fabric solution?

The Nokia Data Center Fabric solution enables you to scale data center networks quickly and easily to meet new business needs while containing costs and improving efficiency.

This next-generation solution brings new levels of openness and automation to data center networking to help your operations teams manage accelerating demand with all the freedom and control they need. We designed the solution for data center network operators and cloud builders across multiple segments, including enterprises, communication service providers (CSPs) webscale companies and cloud service providers such as carrier-neutral network operators, internet exchange providers, data center operators, colocation providers and hosting providers.

The Data Center Fabric solution includes:

  • SR Linux, a uniquely open, extensible and resilient NOS
  • The Fabric Services System, which provides an intent-based automation toolkit to automate all phases of the data center fabric operations life cycle
  • High capacity hardware platforms that enable you to implement modern, massively scalable and reliable data center switching architectures, including the fixed-configuration Nokia 7220 Interconnect Router (IXR) for data center leaf/spine applications, and the modular Nokia 7250 IXR-6e/10e for data center spine and super-spine applications.

GigaOm recognizes Nokia as a data center switching innovation leader and outperformer for a third straight year

Gigaom badges 2024

It’s time to transform your data center switching and cloud networking infrastructures

The pervasive need for all forms of content at the speed of cloud is creating insatiable demand for a new breed of consumer and business services. At the same time, the deployment of new technologies such as 5G and IoT, dominance of AI/ML-based workloads and increasing adoption of hybrid cloud, edge cloud and multicloud approaches are putting massive strain on scale, performance and day-to-day network operations in data center switching and cloud networking infrastructures.

It is becoming clear that data centers need automation to meet these new demands.

Enterprise, CSP and webscale data center network operations teams adopt a do-it-yourself (DIY) or platform-based approach when it comes to automation. Teams that choose DIY embrace the development and integration of in-house or open-source network management and operations tools. The platform approach is for teams that prefer to rely on an out-of-the-box, UI-driven system from a vendor to program and automate tasks across all phases of the network operations life cycle.

Each operational environment is different and has unique needs and problems to solve. Many data center teams see network automation as an answer for making operations simpler and more agile. But the legacy approaches and building blocks associated with current data center switching solutions have kept network automation from going mainstream.

What’s needed is a modern approach to designing, deploying and operating data center fabrics. This approach must use the latest technology advances and best practices to enable support for current and future technologies as CSPs, enterprises and webscale companies move into the cloud-native era and beyond.

Stylized image of woman exploring a growing set of tools for data center fabric transformation.

Accelerate your data center modernization journey with a truly open NOS

The foundation of our Data Center Fabric solution is SR Linux, an open, extensible and resilient NOS designed from the ground up with a management architecture that meets today’s demands for scalability, visibility, cost-efficiency and ease of operations.

A leap forward in innovation

We were committed to designing a solution that would rearchitect the status quo to provide a platform for driving NetOps automation. We also wanted the solution to provide the levels of openness and extensibility you need to thrive in the cloud-native era.

The result is a NOS built on the latest technological innovations, including containers, microservices, open-source projects, model-driven architectures, YANG data models, streaming and “on change” telemetry, along with modern management protocols such as gNMI and REST APIs.

These innovations are combined with highly scalable and advanced routing protocols (e.g., VXLAN, BGP, EVPN) inherited from our proven Service Router Operating System (SR OS) to ensure that you get a feature-rich, hardened, interoperable and secure NOS.

“SR Linux was built with a complete model-driven architecture across all its services, allowing for rich telemetry, which is distributed through modern management protocols like gNMI.”
Scott Brookshire
CTO, OpenColo

OpenColo driving data center innovation

Get started with SR Linux

It’s easy to set up and free to try!

Operational freedom

Our Data Center Fabric solution helps you remove barriers and regain control of your network operations environment so you can focus on your business needs and the applications and capabilities you deliver to your customers.

Tools built into SR Linux enable you to reduce vendor dependence with new levels of extensibility and customizability. For example, you can use our NetOps Development Kit (NDK) to extend your existing platforms with new apps or agents that help solve a particular network deployment or operations problem.

SR Linux is a truly open NOS that is easy to integrate into existing DIY environments. It limits disruption while providing an agile networking solution that meets current and future needs.

You can download a free containerized version of SR Linux from the GitHub container registry. We also offer an open-source tool called Containerlab, which allows you to turn up a customized multivendor data center fabric lab in minutes to test new configurations, or to test drive and learn about SR Linux or a NOS from another vendor.

“Nokia’s data center switch and its network operating system SR Linux met our stringent standards in both innovation and power efficiency. With Nokia’s SR Linux network operating system, we were able to easily integrate into OpenStack Ironic environment to automate our network.”
Rui Gomes
HPC Operations Manager, atNorth

atNorth and Nokia partner to level up sustainable data center networking

Curious to know how your data center switch can be integrated with AI/ML tools like ChatGPT?

With SR Linux, there’s an app for that! Read the blog

400GE/800GE platforms support all roles for data center switching leaf-spine designs

Our family of next-generation data center switching platforms addresses the growing needs of CSPs, enterprises and webscale companies with new levels of flexibility, scalability, speed and resilience.

The Nokia 7250 Interconnect Routers for data center fabrics (7250 IXR-6e/7250 IXR-10e) are modular, high capacity platforms designed to deliver massive scale, reliability and investment protection. They support 400GE, 100GE, 40GE, 25GE and 10GE interfaces for data center super-spine, spine and WAN deployments, and provide an upgrade path to future 800GE speeds without chassis upgrades.

The Nokia 7220 Interconnect Router for data center fabrics (7220 IXR) provides fixed-configuration, high capacity platforms that let you bring unmatched scale, flexibility and operational simplicity to your data center networks and cloud environment. They support 400GE, 100GE, 50GE, 40GE, 25GE, 10GE and 1GE port speeds for data center spine, leaf and management roles.

The Data Center Fabric portfolio is complemented by the flagship Nokia 7750 Service Router (SR) platforms, which can be deployed at the data center edge to support IP backbone connectivity, IP interconnection, IP data center interconnect and IP peering roles.

Switch on the easy button for automating all phases of the operations life cycle

Our Data Center Fabric solution increases network agility by using declarative, intent-based approaches to automate all phases of data center fabric operations—from Day 0 design through Day 1 deployment and Day 2+ operations.

Based on an open Kubernetes framework, our Fabric Services System uses an event-driven, microservice design approach to deliver a true cloud-native operations and automation system. By enabling enhanced NetOps, it gives your operations teams the agility, control and confidence they need to increase the efficiency of the entire data center fabric life cycle.

The Fabric Services System also provides a built-in Digital Sandbox that creates a virtual digital twin of your live production network. It emulates your data center fabric with absolute parity so that you can test and validate network changes in a risk-free environment before you deploy them.

“The vendor ranks high on the NetOps suitability evaluation metric, having designed the Nokia Fabric Services System as an operational toolkit and management system to improve and scale operations across the entire operational life cycle of the fabric. Due to the highly automated design, deployment, and operations processes, Nokia also ranks high on support for Day 0, 1, and 2+ activities.”
Andrew Green
GigaOm Radar for Data Center Switching

Economic benefits of data center network automation

It’s clear that network automation can help reduce effort and cost, but how do you quantify its tangible benefits?

Our business case analysis (BCA) tool—developed by Nokia Bell Labs—answers this question and highlights how much you could save.

You can use the BCA tool to model the job functions, tasks and effort required to design, deploy and operate our Data Center Fabric solution. The BCA tool then uses your input to calculate your effort, power and space cost savings.

The BCA tool uses data modeling to analyze and compare a migration scenario from a present mode of operation (PMO) to two different future mode of operation (FMO) scenarios.

The PMO is based on a current-generation data center networking solution. FMO 1 is a leaf-spine fabric powered by the Nokia SR Linux NOS running on Nokia data center switching platforms based on merchant silicon. FMO 2 extends FMO 1 to include the Nokia Fabric Services System for managing and automating all phases of the data center fabric life cycle—including Day 0 design, Day 1 deployment and Day 2+ operations. This lets you analyze how much time your team could save for each operations task.

Ready to talk?

Please complete the form below.

The form is loading, please wait...