Designing your Data Center for the Future
Hot & Dark – The datacenter of the future
In five years the vast majority of datacenters around today will seem like dinosaurs. You will be able to call up new capacity in 4-6 weeks and you’ll manage your environment to an efficiency level only achievable through the removal of dedicated staffing. The question is, how do you get there and how long will it take you? Your datacenter ownership strategy will be one of the critical keys to your success.
Datacenter Ownership Strategy
I believe there’s good news on the future of the datacenter, but it’s not going to happen by accident. Unfortunately datacenter environments often lag three to six years behind technology improvement opportunities. There are many reasons for this lag, not the least of which are cost, concern over interruption to operations, and assumptions of risk. However, much of these concerns are also buried in the tradition of our “current” thinking relative to datacenter design and our legacy organizational structures.
Without effective ownership, your company will likely continue to lag behind the technology curve. In a fairly traditional organization you might have an IT datacenter manager whose basic responsibilities are to ensure the space is secure, cable management is efficient and that facility’s handles any issues relative to space, power & cooling. Depending on the size of your datacenter the facilities team may be even less focused, there could be 2-4 people that have various duties for datacenter related systems, but there is rarely anyone who owns the datacenter as a system.
My first recommendation to any organization I speak to is to appoint a person to be in charge of the datacenter top-to-bottom (see Data Center Pulse Stack). Once you’ve gotten the organizational focus (including budget) issue taken care of you’re in position to make real change and maintain the operational and efficiency improvements you’ve implemented. Today’s datacenter planners are often caught in the fixed location thinking relative to their facility and I believe now is the time to seriously challenge that way of thinking.
Challenging Conventional Thinking
The datacenter and associated infrastructure of today tend to be very fixed (stuck in place or design). The fixed nature of the infrastructure causes many of us to look for improvement options that fit into the “fixed” box we’ve created for ourselves. This “in the fixed box” thinking is holding us back, but it shouldn’t for much longer. Modularity in datacenter design coupled with portability of environments via virtualization and cloud mean that we have more options than at any time in datacenter history to affect change with minimal disruption. If you’re using the DCP Stack as a guide, you’ll also be in a better position to understand how changes you plan might affect datacenter performance as a whole.
Physical design of your entire building infrastructure is a key to allowing you to build only what you need when you need it. It should also allow you to build different tiers of protection or performance into your facility one containment area at a time and to change it back quickly as necessary.
Integration between the Facility and the Compute Infrastructure
Imagine having your facility alert your technical infrastructure of impending problems so that your compute infrastructure can protect itself by shutting down, moving or just optimizing to reduce load.
In order for the integration of your facility and the compute infrastructure to provide optimal benefit you’re going to need a “portable” solution. With portability of the compute infrastructure (network, storage & applications) you can distribute your load and provide redundancy and protection for your work load. There are a number of virtualization tools that can provide you with some level of portability. You should also being looking to your partners in the network space for their progress on portability.
In summary if you’ve designed your datacenter of the future appropriately, you will probably have some or all of the following:
- A distributed, but highly efficient datacenter footprint that is dark most of the time.
- Compute infrastructure that is portable and whose individual components make up capacity, not capability, which allows for fail in place and replacement on a schedule.
- Low cost of ownership through efficient space acquisition (you should no longer be buying datacenter space that costs you money for years before being put into use).
- Removal of forced air cooling, including potentially replacing server fans with liquid cooling.
- Greatly improved flexibility for building what you need, in the quantity you need, when you need it.
Most importantly you will be providing your customer with a lower cost of ownership, higher level of availability, faster time to market and helping the environment.