Data Center Deployment Tips on a Global Scale | Corning

Tips for Managing Data Center Deployment on a Global Scale

Tips for Managing Data Center Deployment on a Global Scale

Whether you’re a global firm or a more local enterprise, we believe many of these best practices for data center deployment can apply to your organization.

At Corning, we’re privileged to work with thousands of customers on their data centers, providing support ranging from design advice and installation support to full-project management. And we learn a lot from them about the best ways to handle for multilocation deployment.

Whether you’re a global firm with networks in dozens of countries, or a more local enterprise with sites in the same time zone, we believe many of these best practices apply to your organization, too.

Standardize on a Consistent Infrastructure

It can be tempting to let each site design a data center cabling infrastructure solely on its unique needs or physical location. But this approach often hinders scalability within the organization. For example, staff members moving from one location to another may not be able to get up to speed quickly on the new data center physical topography because of variations from their last location. It can be difficult to implement organization-wide network moves, adds, and changes (MACs) on a predictable, timely, and manageable scale when each location has its own infrastructure platform.

Of course, many of our customers also recognize that a degree of locational flexibility is critical. As many organizations aren’t in a position to construct every building identically from the foundation up, you’ll inevitably end up with physical differences between sites. One may require hardware that is shallow due to physical space limitations – in the other location it may make more sense to increase density per square ft to address the space issue. In some of our customers’ global locations, labor challenges make a factory-terminated solution the only option; yet in another country, field-termination is the smarter choice.

With this in mind, consider solutions that allow for location-specific needs while still leveraging components that work together. We believe our EDGE™ solutions data center platform offers big advantages through multiple “versions” that are still interoperable. Our platform also allows easy, organization-wide stocking of components. Your procurement teams will appreciate the purchasing efficiencies this creates.

Make Sure Global Really Means Global

When you’re evaluating your infrastructure options, look carefully at where products are manufactured. Sure, today’s products can be shipped anywhere, but not necessarily cheaply or quickly. Manufacturers who produce in your region – or at a minimum, keep the technology stocked with local distributors – are more likely to get products in your hands on time and on budget.

On a related note, check to see if the same product is manufactured in two different locations. If you’re getting multiple shipments, you might not be getting the same product. Sometimes the specifications, materials, and even dimensions of a product may differ from location to location. Truly global solutions will ensure that the technology you select to deploy across your organization is identical from site to site.

While this may seem to be only a concern for those organizations that really are global, close-to-home operations can take a lesson from this as well. Some of our customers, to their eventual regret, have treated base components – like cables or connectors – between manufacturers as interchangeable commodities. This can certainly result in performance differences. More often, it leads to installation and maintenance practices that differ by location, also creating scalability inefficiencies for your organization in the future.

Create a Global – or Organization-Wide – Implementation Team

Many of our customers have built “tiger teams” to deploy their data centers, at initial installation and/or with large MACs. They travel to each location, for weeks or months at a time, building the location to the global requirement, training the local team and then moving to the next location. While the travel and boarding expenses of these teams might seem unattractive compared to using on-site teams, our customers who choose this method say the cost is more than made up for by the speed of deployment gained. These extremely practiced teams have little to no learning curve as they reach each new location. They also cite the consistency of the deployment across the organization’s locations as another advantage, believing that expert installation lessens the risk of future downtime or performance issues.

Some technology manufacturers, like Corning, have project management or field installation services for customers who do not have the resources to assemble an organization-wide tiger team of their own.