Operating in colocation data centers is not a “new” concept by any means. Colo services have been viable options for companies for more than a decade. What is new is the rate of acceleration organizations are moving their production and data redundancy (DR) to colo.
While moving to colo is common-practice for many, it hasn’t been so for some until recently, and they’re now seeing these services as viable options for their businesses for the first time. It is important that these newcomers recognize not only the biggest but also the most common operational mistakes IT organizations make when moving their IT environments into colocation.
Choosing the Wrong Colo Provider
First and foremost, a data center must support the technology that ultimately supports the business goals and objectives. No two colocation data center providers are the same, so clearly identifying the key strength of a colo and aligning that with the data center business goals and objectives should guide selection of the provider.
For example, an organization’s goal may be to improve disaster recovery, so the objective would be to have its on-premises data center environment duplicated in a colo. In this scenario, you want the colo far enough from the on-prem facility so that both cannot be brought down by a single storm and far enough so that the two facilities are on different power grids.
Another goal may be to reduce future capital expenditures by moving all IT equipment into a colocation data center instead of building a new server room or upgrading an existing one. This may require that the colo charge monthly rent without any upfront build-out costs.
Not Knowing If the Colo Is “On the Hook”
With any agreement, it is important to have a clear understanding of the “fine print”-- especially when it comes to the service level agreements.
A colo can write its SLAs in a way that gives it a lot of leeway if outages occur. For instance, do they allow themselves a specific number of planned outages throughout a given year? If yes, this colo offers significantly less uptime than one that guarantees uninterrupted uptime regardless of whether the outage is planned or unplanned. Less than 100 percent uptime may be acceptable for DR purposes, but probably not for mission-critical production equipment.
Additionally, you should be aware if there is a termination clause that can be triggered if the colocation data center provider repeatedly violates its SLA. Termination clauses allow tenants to terminate the contract and move out if they become fed up. The quality of service and anticipated uptime can vary between each colo based on what is stated in their service level agreements. The fine print indicates quality of service and amount of uptime a colo customer can expect to receive.
Other “fine print” areas to examine are:
· Does the colo only address power in their SLA, or include network uptime and/or promise cooling and humidity to be within and a certain range?
· Does the colo quantify the amount of credit that will be given if SLA terms are not met?
Getting Blindsided by Hidden Costs
No two colocation data center providers charge for services the same way. One colo may charge initial fees to install server cabinets, power, etc., while another may amortize these costs over the length of the lease. As a result, comparing prices between each colo can be mystifying.
The best way to successfully compare prices is by projecting future growth in terms of type and quantity of hardware needed in years to come. This hardware will dictate space, power, and connectivity needed which are then used by colos to create their prices. In doing so, an organization should also create a list of all services they might need because each service may come with a fee.
In addition to the hidden costs that a colo may charge for space and power, network connectivity should also be considered. Which communications carrier an organization currently uses, and which carriers already reside at a particular colo will affect the carrier’s cost. If an organization’s carrier does not currently reside in the colo, there may be a cost to have the carrier connect to that colo. An organization’s business continuity requirements also need to be considered, as this will dictate the number of diverse carrier routes needed between the colo and the organization’s headquarters and other satellite locations.
Once growth projections are estimated and all needed services are determined, an organization can then request pricing based on space, power, connectivity, and services. Furthermore, all pricing should be forecasted several years into the future. This provides the organization with an apples-to-apples comparison for total cost of ownership between all colocation data centers being considered. Without this cost information an organization may move into a colo that seems like a good deal only to find unbudgeted-for hidden costs later on.
Underestimating the Move-In Timeline
Once a contract is signed, colocation providers can often meet very tight deadlines to have server cabinets in place, power strips energized, and your team equipped with security badges. However, long-lead-time items beyond a tenant’s control may prevent them from moving in.
One very common lag item is the activation of a carrier circuit. While all other items can be turned around in less than a month, carrier circuits often take at least 90 days. Without carrier connectivity, officially migrating to a colo is impossible.
To avoid timeline delays, it is important that the timeline developed and agreed to by the provider anticipates and takes into account all aspects, not just when the “movers” can get the cabinets installed.
The providers are experts at the move-in phase of the process. Don’t try to do it all yourself; engage the colo in discussing the requirements – from special constraints to security – before the first cabinet is packed.
As with any data center strategy, the decisions that need to be made must be grounded in business goals and objectives. With a solid understanding of what is driving the decisions, you can ask and answer many of the fundamental questions that, when overlooked, can cause issues throughout your relationship with the colo provider. Colocation data centers are a viable and appealing option for any data center operation – as long as you avoid making these common mistakes.
About the author: Tim Kittila is Parallel Technologies’ Director of Data Center Strategy. He oversees the company’s data center consulting and services to help companies with their data center, whether it is a privately-owned data center, colocation facility or a combination of the two. Earlier in his career at Parallel, Kittila served as Director of Data Center Infrastructure Strategy and was responsible for data center design/build solutions and led the mechanical and electrical data center practice, including engineering assessments, design-build, construction project management and environmental monitoring. Before joining Parallel Technologies in 2010, he was vice president at Hypertect, a data center infrastructure company. Kittila earned his bachelor of science in mechanical engineering from Virginia Tech and holds a master’s degree in business from the University of Delaware’s Lerner School of Business.