IT Focus Area: infrastructure operations
January 6, 2016
Why You Should Think Outside of the Box for Data Center Migrations
Editor's Note: Sirius and Forsythe are now one company. Sirius acquired Forsythe in October 2017 and we are pleased to share their exceptional thought leadership with you.
When most of people think of the data center, the image that comes to mind is a locked room tucked away in a building filled with racks of servers, connected together by miles of spaghetti-like cables.
It is an image that has been popularized by movies and TV shows where the hero or villain needs to tap into a company’s data in order to accomplish something central to the plot.
Yet increasingly, if someone like Ethan Hunt in Mission: Impossible were to go to the trouble of breaking into a company to get to the data center, he might discover that the most impossible part of the mission is determining where the application and data of interest reside. There is a good chance the company itself would also have trouble mapping data and functionality to assets within their own data center.
Virtualization, cloud offerings, and “out of the box” methods of hosting applications and data have made data center migrations much more challenging today.
Rather than moving assets from one location to another, there are a wide range of options available today that make data center migrations considerably more complex.
So why go to the trouble of moving data centers?
There can be any number of reasons to perform a data center migration, including physical (e.g. power and cooling), increased need for resiliency or uptime, corporate acquisitions requiring asset consolidation or more proactive data center strategy initiatives.
Whatever the driver, it is important to understand all the destination options and data center migration methodologies to ensure you make the best choices so that not only does your migration project go smoothly, but you are taking advantage (technically and/or financially) of the inherent disruption.
How Data Center Migration Has Changed
Until recently, computing at the enterprise level has been focused on the hardware (i.e. the box). A company needed enough servers and storage in the data center to accomplish what the business needed. When the facility no longer supported the “boxes” serving that purpose well enough, it became time to migrate to a new environment.
As a result, when a company migrated their data center from one location to another, organizations were essentially concerned with boxes. The process involved moving the existing infrastructure to a new data center with increased capacity. More recently, companies tried to refresh to upgraded hardware in order to gain speed, increase storage, improve uptime, and build for the future as part of the migration.
It was more of a move “as is” upgrade process, similar to moving files off an old personal computer (PC) and onto a new one—only on a much larger scale. And just like with a PC, the hardware you were moving to was central to the success of the data center migration.
Today, the definition of data center has expanded beyond the four walls of the enterprise. Many enterprises have shifted toward a hybrid data center model. In today’s modern data center, applications and data can reside in many different places—private cloud, public cloud, legacy systems, a colocation facility, infrastructure as a service (IaaS), software as a service (SaaS), a virtualized environment, or (in most instances) a combination of these options.
Increasing awareness has been appropriately placed on the importance of application performance. Where the infrastructure resides is less important than the ability to deliver the performance required and meet service level agreements (SLAs).
Preparing for a Migration
Much like the physical moves of old, proper preparation and planning is the key for a successful migration. The best practices approach to starting the project called for understanding everything in the existing environment and what each component did. There was no substitute then, as now, for a thorough asset analysis and application mapping.
Today's most successful migrations start with a focus on the criticality of the applications the data center supports, how they are affected by downtime, what sort of downtime tolerance they have and how you can mitigate it. For example, if you are a financial institution offering 24/7 online banking, downtime tolerance is essentially zero. In performing a data center migration, you have to plan the actual migration and choose your target with eliminating downtime as one of the top priorities.
The Steep Learning Curve
For most organizations, a full data center migration is not an every-day occurrence but rather an every few years occurrence. That means when the time comes to begin planning, companies often face a steep learning curve, both in terms of understanding the various options available to them and how to determine which suit their organizations best.
Most companies then have to learn the tools and best practices at a deep level in order to give the migration the best chance for success. Since they will be doing all of this on top of their day-to-day duties—which they are probably already struggling to fulfill—it seems like a lot of work to learn something that, once it is completed, will not be needed again for a few years. This is where it may be helpful to gain an outside perspective from a partner that specializes in data center migrations.
A company with expertise in data center migration already knows the options, the tools and the best practices. They bring a wealth of experience from organizations similar to yours, and can help you see the red flags and avoid the hidden “gotchas” that normally don’t show up until you are elbows-deep into the project. They can apply that experience to help you determine the required and desired benefits, look at the impact your decisions will have downstream, and then create a data center migration plan that will produce the desired results with minimal disruption, while maximizing peripheral benefits available during migrations.
They will also help you work across silos within your own organization to ensure all necessary voices are heard, in the sequence required, as the plan is developed and implemented. It is often difficult for business and IT leaders to think outside their own “boxes” of responsibility. A partner can act as a mediator and sounding board, bringing all of the parties together in one place so you have a more accurate “big picture” view. That alone may help you eliminate some of the data center migration options, keeping you from making critical missteps that can have serious consequences later on.
Along with all that, a good partner will build quality into the process rather than inspecting it in at the end. Again, this is where their experience comes into play teaming with your organization. The better job you do mapping applications and making decisions up-front, the smoother the migration will go, the less risk of unplanned downtime your company will have, and the better-positioned you will be to deliver real business value to the organization faster.
Forget the Box
The term data center today is almost a misnomer. It implies a self-contained box where data resides. But, in the modern enterprise, applications and data can reside in any number of places, each of which serves a particular purpose in the delivery of information to business users. While that may make it easier to use, it also creates new challenges when you want to migrate from your current “box” to one (or more) of these new options.
Working with a partner that already knows how to think outside the data center box can be invaluable when it comes to the planning and execution of this type of big migration project. It will save you the extra work of having to become an instant expert, put you on the fast track to success and avoid those red flags.