The New York Times misses the big picture with its take on data centers as "wildcat" utilities that need government oversight. There is no data center monopoly in northern N.J.
A "wildcat power broker" in this position, if I understand what Glanz means, might use the scarcity of power as a means to increase prices at will, or worse, help generate the shortage in the first place. The northern N.J. data centers are not power brokers, in that sense. They are subject to the limits of what their supplier and the grid can deliver to them. Their control is over a set of resources that defines a data center -- the network connections, the secure space, the cooling system -- with electricity as one element of that offering.
If they charge double for power, it's possible that's because they have signed contracts to deliver twice the amount the customer typically uses and must bear the expense of keeping that supply available, even if the customer isn't using it 85% of the time. Failing to do so could lead to financial penalties written into a customer's service level agreement. While there may be some calculated risks undertaken by the data center owner in allocating power, the reckless reselling of power would surely lead to a day of reckoning.
In addition to having to supply direct power, the data center operator must also have an uninterruptible power supply installed, in case of grid delivery failure. The backup must kick in the instant the grid fails. Otherwise data will be lost in running applications and data integrity will be compromised as systems try to come back online. In some cases, that's a bank of 12-volt batteries siphoning off a trickle charge from the incoming power supply. Today it's usually a more sophisticated mechanism that attempts to minimize the power lost, such as Vantage's insulated gate bipolar transistor unit. In this case, it's good to be bipolar; it cuts resistance to the electricity supply's flow and reduces power loss.
In addition, for space to be suitable as a data center, it must have emergency backup generators capable of replacing all the power that's normally imported. Perhaps Glanz knows how to amortize the expense of insulated gate uninterruptible power supplies and backup systems, but I suspect most public utility commissions do not. (But the N.J. utility commissioners know quite well the value of backup systems in the wake of a storm, such as Hurricane Sandy.)
A wholesale data center is not a wildcat power broker with the goal of profiteering from a stranglehold on the power supply. It's a combination of services in a location where such a set is in high demand, such as outside New York in close proximity to the trading exchanges. A public utility, seeing potential business at risk, can contract for more power, add substations, upgrade its grid. The wholesale data center cannot. Power is brokered to it, but it cannot become a power broker with the ability to, at will, expand or contract the supply.
"The sale and resale of power is subject to a welter of regulations and price controls ... But the capacity pricing by data centers ... appears not to have registered with utility regulators," Glanz concludes. In this case, the absence of regulation is a good thing.
There is no data center monopoly in northern N.J. On the contrary, it is one of the most competitive regions for building data center space in the world. Let the data center operators run their business as they see fit. They're the ones who know it best. And let's hope the regulators who read The New York Times keep their common sense and don't take Glanz's recommendations too much to heart.
Multicloud Infrastructure & Application ManagementEnterprise cloud adoption has evolved to the point where hybrid public/private cloud designs and use of multiple providers is common. Who among us has mastered provisioning resources in different clouds; allocating the right resources to each application; assigning applications to the "best" cloud provider based on performance or reliability requirements.
InformationWeek Tech Digest, Nov. 10, 2014Just 30% of respondents to our new survey say their companies are very or extremely effective at identifying critical data and analyzing it to make decisions, down from 42% in 2013. What gives?