FOCUS POWER


Issue 8, Feb/Mar 10


centers can contribute electricity to it. One of the ways would be to contribute electricity produced during generator tests that are run periodically – something Mark Bramfitt, the former data center energy efficiency czar at PG&E (a California utility) has proposed at many industry events.


Van Helten’s third theoretical connection point between the development of a smart grid and the data center sector is the main function of the data center: the transport, processing and storage of digital data.


With more and more chip-bearing devices coming online, collecting information about building energy use and sending it to the grid operator, that operator has an enormous amount of data to process and store. Data will be moving in all directions among smart meters, energy sources and the grid itself.


Van Helten says it is likely that utilities, instead of investing into building out and managing their own data centers, could start outsourcing that task to professional data center operators. “If you want to build that infrastructure for yourself, that’s pretty expensive.”


Along with the need to process, store and transport all the data the grid collects is a need to secure that data – also something data center operators specialise in. “A lot of the data centers are pretty secure,” Van Helten says. “That’s a Fort Knox for information.”


DELIVERING ON PROMISES Ronald Bowman, executive vice president at Tishman Technologies, says a bit of skepticism is needed in the current discussion of smart grids – an expression he says had been “tortured” by politicians and the media to the point of having little meaning.


Bowman says a lot of “intellectual dishonesty” took place in the market about the costs and benefits of smart grids on the part of vendors of products that enable it. “You need to level the expectation of what the grid can actually do for you,” Bowman says.


Implementing a smart grid is “remarkably expensive” and a lot of what it does has already been done manually through the use of common communication methods.


A smart grid would, in effect, enable grid management processes that already take place to be highly automated.


Rather than becoming integrated with 22 www.datacenterdynamics.com


smart grids, Bowman says data centers in the US would be better off moving toward increased independence of the electrical grid by increasing their energy efficiency and developing their own generation capacity, or building facilities next to generation plants.


He says that because today’s electrical grid in the US is unreliable, “you’d want to go with co-generation, or bolt on your data center closer to a power-creation source”.


Bringing a facility closer to the generation source reduces the risk of power disruption caused by grid failure.


Some organisations have made great headway with co-generation. Bowman uses the well-publicised Syracuse University data center project as an example. Along with IBM, the university is building a data center that will be powered completely by an on-site co-generation system, which will be based on a micro turbine that is fuelled by natural gas.


“On-site co-generation absolutely makes sense and should be done more,” Bowman says. “Self-help is a far greater silver bullet than putting a meter on every device in the country and calling it smart.”


MORE CONCRETE DATA A key function of a smart grid is enabling effective demand response. This is when electricity consumers adjust their consumption based on the load on the grid. If the grid is at peak demand, consumers reduce their consumption to ease the strain.


One of the most recent research projects funded by the DoE is a study by Lawrence Berkeley National Laboratories, in collaboration with the California Energy Commission, on the feasibility of implementing demand response in data centers. This was a scoping study – a study that identifies opportunities and makes recommendations for further exploration.


PG&E contributed funding to the project. As DCD Focus went to press, the final report on the study was completed but had not yet been published.


Lawrence Berkeley National Laboratory researcher Girish Ghatikar, one of the study’s authors, provided DatacenterDynamics with insight into some of the project’s conclusions.


The researchers concluded that there was significant potential for demand response in data centers. The most likely candidates


for early adoption, however, would be data centers that support non-mission-critical applications, such as computer rooms that serve research labs.


One opportunity for demand response and load reduction was using virtualisation to reduce energy consumption by IT equipment, according to the researchers.


There are also opportunities for demand response on the facilities side: for example, HVAC and lighting systems.


Virtualisation is a promising tool in implementing demand response because it can be used to dynamically control utilisation of server processors and thus the servers’ overall electricity consumption.


One of the study’s key findings is that a quicker demand response and better planning is most likely to be achieved if both IT and facility infrastructure work together to shed load. If IT load is reduced, so is demand for cooling. Along with a reduced load, losses associated with power distribution are reduced as well.


Researchers say that more in-depth studies are needed in order to quantify the benefits that participation in utility demand response programmes would bring data center operators. Quantification of the effects that such participation would have on data center performance, quality of service and the life span of equipment are also needed.


AT A CROSSROADS With the idea of the smart grid still remaining little more than just an idea, and with all the uncertainty surrounding it, works in progress and convoluted messages, the industry is facing yet another decision that could potentially have radical implications for its future: should data centers participate in the development of the smart grid, being such crucial enablers of the concept? Or, is it time they considered alternatives – such as Bowman’s advice – and started thinking about getting off the grid altogether instead? 


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52