Chris Curtis is the co-founder and SVP of Development for Compass Datacenters. We are publishing a series of posts from Chris that will take you inside the complexity of the construction process for data centers. He will explore the ups and downs (and mud and rain) of constructing data center facilities and the creative problem-solving required for the unexpected issues that sometimes arise with every construction process. For more, see Chris' previous column on the start of construction and previous columns on the planning process.
When a customer purchases a data center, they specify the date that they would like to take possession of it. They tend to be pretty insistent on these things. I’ve yet to read a lease or purchase agreement in which terminology like “mid-month, “June-ish” or “whenever it’s done” has been used to designate the time that the project must be completed.
From a developer’s standpoint, this means that the project must have a schedule and that schedule must allow for the fact that during a six-month construction period, you’re probably going to run into some less than optimal weather. Of course, these weather considerations tend to vary with geography--you don’t build in many snow days for a data center in Phoenix, for example—you still have to take them into account. In other words: when developing a realistic development schedule, “weather days” must be built in.
Because I’ve learned some hard lessons about the varieties of weather, and their associated schedule impact, I looked at the average rain volume for the project’s location and built the estimated “weather days” into the schedule. Unfortunately, my calculations did not anticipate that, over the course of the project, the area would experience a volume of precipitation that can only be described as “biblical” in nature.
In my case, rain completely stopped work for 45 days, and impacted a total of 69 days. In other words, 38 percent of our already aggressive schedule was blown away by the weather. At various times, the customer’s future data center appeared to be surrounded by a moat. Good if you’re building an impregnable medieval fortress, but for the future home of a million or so dollars of computing gear—not so much.
Weather Impacts, Schedule Struggles
As you might expect, rain and construction are not a productive combination. People are cranky. Imagine trying to your job if you were sitting in the equivalent of a cold shower all day. I’ll bet you wouldn’t be a barrel of laughs if you had to type on your keyboard with cold, "pruny" fingers for eight hours or so, electricians tend to be especially averse to working in the rain with water being a conductor of electricity and all. I guess the heightened potential of a few thousand volts coursing through your body does tend to give you a little different perspective on the term “work-related accident.”
Things tend to progress a little slower. Concrete takes longer to dry, trucks get stuck in the mud, and the humidity wreaks havoc on the painting process. The net result of this data center version of Noah’s Ark is that all of the slack in our schedule is now gone, or more accurately, has been gone for a while.
Project Management Needed, And Some Persuasion
A project schedule is a funny thing. Although everyone understands that you need one, it takes a strong project management to get sub-contractors to deliver as committed. I guess seeing everything you’ve committed to, written down on paper, turns all the phrases that you used to get the business like “piece of cake” or “no problem” into one giant “holy crap.”
Based on this level of initial enthusiasm, you can only imagine the level of sheer panic on behalf of all involved when that same schedule has to be compressed to meet the deadline. Situations like this test the mettle of even the most battle hardened data center developer. Resistance is fierce, excuses are made, and there is always some begging for extra time. Some will even refuse to move forward. But you still have to find a way to gain their compliance to meet the date and maintain quality. I don’t know what methods other developers use to overcome these issues, but I rely on a combination of coordination, communication, persuasion and escalation to senior executives.
At the end of the day, however, holding their money until successful completion always proves to be the most effective technique. Okay, so I’m no Vince Lombardi, or Tony Robbins for that matter, but for all you new developers out there I say that, “When the going gets rough, pull out the biggest stick in your bag.”
Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.