Tips for migrating to a software-defined data center / centre

• Print

Details

Published: Wednesday, 17 February 2016 08:42

Not only does a software-defined data center / centre provide CIOs with a way to improve disaster recovery and combat shadow IT, but it is also the first step in moving toward a truly service-defined enterprise and acts as the catalyst that redefines the CIO’s role from technologist to business enabler and innovator.

A software-defined data center / centre (SDDC) is a data center in which all of the components – network, compute, storage – are virtualized and controllable from a single unified interface.  By creating a management layer that spans all data center resources, IT departments can more quickly and easily define, develop and deploy services in response to an organization’s rapidly changing business needs.  IT tasks that once took months to complete can literally become minute-long, single-click exercises, giving the IT department a new way to keep pace with line-of-business service requests. 

To assist organizations thinking of making the move to SDDC, Logicalis US has provided the following tips:

1. Start with a plan: begin by assessing where the company is now as well as where it wants to go.  One of the biggest reasons SDDC projects fail is because the company is hyper-focused on the technological end goal to the exclusion of other factors.  True, SDDC is a technological change, but it’s not just about the physical assets in the data center.  There are people and process components involved in making a move to SDDC successful; to succeed, the organization has to be ready for these changes, and everyone concerned must be on board.

2. Virtualize where possible: create virtualized containers on top of all computing devices; by creating a software layer that’s separate from the physical hardware, IT can take action by building, configuring, changing and decommissioning hardware all from one unified management interface.  Dynamic swaps of space and configurations can be done on the fly; IT is no longer bound by how much capacity is available at a given point in time.  Imagine it this way: If you buried all your money in the back yard in various containers holding different denominations, you would have to dig each container up each time you needed a specific sum rather than simply going to the bank and withdrawing exactly what you need when you need it most. That’s the difference between managing a physical data center and a virtualized one.

3. Buy the right fuel: think of compute capacity as the fuel that runs the data center.  Like gasoline for cars, compute capacity comes in different varieties: public cloud, on-premise private cloud, or a hybrid combination of the two.  The goal is to buy the compute capacity that will most cost effectively fuel the engine of the organization’s new software-defined data center.

4. Orchestrate and automate: CIOs must ask themselves how much of what they do on a daily basis is held strictly in their heads or resides inside the minds of key employees?  If some of those tasks are truly repetitive, they can be orchestrated and automated, freeing IT staff from mundane processes by moving that intellectual property into software and, thereby, allowing internal talent to focus on innovation and market differentiators that create true competitive advantage.

5. Create a service catalog / catalogue: Typical IT departments have key services they provide over and over again.  They are prompted to provide these services when they receive a service ticket, an email, a phone call, or have someone yelling at them – but there is a better way.  Just like retailers offering everything from sporting goods to men’s wear, IT departments can create a catalog of technology services that allows people to simply place an order for the service they need and be charged for it internally.  The idea of providing technology as a service to internal users via a catalog or portal also gives IT the ability to illustrate time, value and return on investment in a clearer and more meaningful way.  However, a common pitfall is the temptation to focus more time and attention on creating a beautiful service catalog than on bolstering the services and IT functions behind it.  If the CEO uses the portal only to discover that the system doesn’t work properly, the CIO and IT department may lose more than an opportunity to impress.

6. Consider disaster recovery: have a mobile phone with a broken screen?  No problem. As quickly as you can get to the carrier’s store, you can buy a new phone and simply move your contacts, photos and apps over to the new one; that’s because those valuable data sets and applications are software-based.  The same holds true in a software-defined data center. When everything is virtual, disaster recovery becomes markedly easier.  No more hours and weeks of building a complicated DR plan; since the data center is all operating from software, if the hardware goes down, just replace the hardware and copy the software onto it again.