You think you have it tough. Imagine going to the CFO and asking for money to build a corporate facility to meet Nuclear Regulatory Commission design criteria. A structure with 8-foot-thick walls. One designed to continue to operate even if a tornado hit head-on.
In other words, a high-reliability data center.
Welcome to the world of Christopher A. Wade, manager, facilities/critical power in the WalMart information systems division. It's a world where the level of redundancy must be balanced against the availability of a mirrored site. A place where 100-year flood plains matter. Where more uptime sounds better to everyone - until they hear the price tag.
And where the reliability of the corporate network may depend to a considerable degree on the quality of the facility executive's network of flesh and blood contacts.
That's not to say numbers don't matter. But consider one question the CFO is sure to ask about a proposed facility: Why isn't the old one good enough any more?
"You have to be able to tell the financial people what the problem is," Wade says. "And it has to be a problem to them. They want to know what the consequences to the business will be if action isn't taken. How much money will we lose?"
That's not a question the facility executive can answer singlehandedly. "Our role in facilities is to provide the facility infrastructure design/build costs," he says. To figure out how much downtime will cost the organization, the facility executive has to tap - or build - an internal network of operating departments.
"I talk to them about scenarios," he says. "If we lose this equipment, what's the impact on the business?"
In those conversations, it's not enough to say that a facility failure will mean the loss of telecommunications capability in a particular area. Instead, the consequences must be put in terms of orders that can't be shipped or credit card transactions that can't be processed. "You have to keep talking until you come to some money," Wade says.
Finding the money is only one example of the networking Wade does before he walks into the CFO's office. He goes to the users of space, for example, to understand their needs. And he assembles a committee of key players from functions like structured cabling, network engineering and telecommunications to identify alternatives, costs, benefits and potential problems.
A LOT TO KNOW
Even as he gathers information from within the company, Wade is calling on another network - facility executives responsible for similar data centers, along with architects and engineers who design data centers - to help determine the life cycle of various data center options.
There's also a wealth of material from the biggest network of all: the Web. "I'm on the Internet all the time," he says. And he turns to yet another sort of network - the one that holds the database for facility maintenance and operating costs - to evaluate and justify recommendations for building materials and systems.
Those benchmarks, whether drawn from internal records or from outside sources, are invaluable. "It's some of the best information you can have," he says.
Armed with documentation gleaned from an array of sources, Wade is ready to make a proposal - or rather, his proposals.
"You have to give them options," he says. "If they don't have any alternatives, an idea is hard to sell." He likes to go into a meeting with three choices - plus the option of doing nothing. That's when the questions begin. What's the business advantage of one level of reliability over another? Why select a new generation of building automation over an old one? At that point, it's crucial to have, not only answers, but sources - "something to back you up," as Wade says.
That's because, when it comes to preparing a presentation for the CFO, the place to start isn't always with numbers. It's often with people. And that point is as valid for a small lighting retrofit as it is for the construction of a five 9s data center.
No comments:
Post a Comment