Skip to main content

Building your business case for a new data center switching fabric

Building your business case for a new data center switching fabric

Automation is a journey for most of our data center operator customers. They start with the simple things: building users, adding interfaces, or even doing configuration of simple services in their environment. They all want to add in additional automations, but what they've told us is that the technology that they chose for their last generation of data center network deployments doesn't meet the need of their new operational model.

Their DevOps teams need more flexibility, and they want provisioning to be done faster, which presses their operations team to safely add more configuration into the network. They need the right tools, they need the right process, and they need additional automations by vendors like us to help them achieve those goals.

We have the data center fabric solution that can help them meet these goals. The solution includes, SR Linux, an open, extensible and programmable Network Operating System (NOS) that improves operational agility and flexibility, a broad portfolio of high-performance data center hardware platforms and the Fabric Services System, a fabric operations and automation platform which uses cloud-native and intent-based approaches and complements SR Linux to delivers a modern and class leading NetOps toolkit.

  • Our SR Linux and Fabric Services System bring data center teams the benefits they are looking for including: Increased reliability and availability of application workloads
  • Scalable, simplified operations that keep pace with business growth while efficiently managing resources
  • Automation of manual operations tasks that are error prone, repetitive or complex to perform
  • Reduced operational risk when making frequent configuration adds, moves and changes to meet evolving application needs
  • True network emulation that reduces the need for physical test lab equipment
  • Translation of raw data (telemetry) into actionable insights, which help perform corrective actions, as required and with the desired level of automation
  • Improved operational metrics such as uptime, performance, mean time to repair and mean time to innocence.

In our meetings with customers the question that has been hard to demonstrate is how to migrate from their present mode of operation (PMO) to a future mode of operation (FMO) that cost-effectively meets their needs. This difficulty in quantifying the operational savings and benefits led us to develop an analytic tool to build a business case for adopting our data center fabric.

We worked closely with our Bell Labs Consulting colleagues on a data center fabric Business Case Analysis (BCA) along with an online tool to help operations teams understand what the trade-offs are for different architectural choices and decisions that they may make when deploying their next data center.

We designed two different scenarios and compared them. The present mode of operation (PMO) generically represents where customers are today. It is a leaf-spine fabric design that might be used by a customer who's looking to continue with their existing technology for their next data center build. We model the leaf-spine fabric based on inputs from the operator about their resource allocations per job function and for their associated job tasks. We've included all the necessary components, both hardware as well as software. There is some basic automation that has been put into the PMO network model, but it is not significant, maybe the Ansible level of orchestration of configuration, some stats collection and reporting, as well as alarming, but not advanced automation.

The future mode of operation (FMO) is based on the Nokia's data center fabric solution. It includes inputs for the size of the fabric, as well as resource allocation for operational job functions and tasks, for all three phases of the data center fabric operations lifecycle. Combining this analysis of both the architecture and physical aspects of the design with job tasks provides a complete and comprehensive view of what it's going to cost operationally to manage a new design.

The Nokia BCA offers a true indicator of cumulative effort savings because it models a comprehensive set of job functions and associated tasks across all phases of the fabric operations life cycle. Existing studies often focus on CAPEX savings and those that do include OPEX savings often only cover a subset of the operations phases, job functions and tasks.

Cumulative savings over 4 years across all operations phases and job tasks

Besides modelling the benefits of enhanced operations for operations job functions and tasks, the BCA also calculates power and space savings and capacity increases that can be achieved when moving to a higher scale next-generation leaf-spice fabric architecture. These include power savings of up to 49 percent and space savings of up to 56 percent with a 2.3x capacity increase for Gigabit Ethernet ports. This information can be helpful as a data point in internal business cases covering CAPEX spend associated with future data center buildouts.

If you are interested and need some help making your internal business case, you can access the BCA online tool and take it for a test drive to see what savings you could potentially see from leveraging Nokia technology for your next data center deployment. And we’re always happy to work with you to customize the analysis further if you need.


Jonathon Lundstrom

About Jonathon Lundstrom

With 20 years in telecommunication, Jon’s career has spanned two companies (Nortel, Alcatel-Lucent/Nokia), five key technology revolutions (IP, MPLS, SDN, NFV, NetOps/DevOps), and dozens of network deployments with carriers and enterprises of all sizes.

Connect with Jon on LinkedIn

Article tags