Dell, NEI Team Up on Servers for DC-Powered Data Centers

Dell is getting a new line of servers ready for the coming world of DC-powered data centers.

Dell will work with NEI to manufacture and market the servers, based on Dell's PowerEdge R710 platform, the two companies announced Tuesday. The move puts Dell alongside competitors such as Hewlett-Packard and Sun Microsystems that now have DC-powered servers on the market.

Electricity grids deliver power in alternating current, or AC. But servers and other computer equipment use direct current, or DC. Converting AC to DC is mostly done with power converters for each individual device or server rack.

But powering an entire data center with DC power could save the inefficiencies associated with this method, according to Lawrence Berkeley National Laboratory. That includes reducing the inefficiencies in converting incoming grid AC to DC, and then back again, several times to filter out grid power glitches, which many data centers do (see DC For Data Centers?)

"You get much more efficient power usage by making them DC across the board," said Jeff Hudgins, NEI's vice president of marketing. "It's a growing trend."

The telecommunication industry has used central DC power for some time, he added. Japan's NTT DoCoMo unveiled a DC-powered data center in February, one it said could cut energy use significantly, with a goal of a 50 percent reduction in associated greenhouse-gas emissions (see NTT DoCoMo Tests DC-Powered Data Center).

Consider it one of many ways data centers are trying to squeeze energy efficiency out of their operations. Data centers consume about 1.5 percent of the electricity in the United States, but that use is set to double by 2012, requiring the equivalent of 10 new power plants, according to an EPA study (see Data Centers Could Hit 'Resource Crisis').

That's a problem for data centers that are facing rising power bills, or may be unable to expand because they've maxed out the power available from their utility's local distribution grid. It's also a problem for utilities, some of which are offering incentives to data centers that can show efficiency improvements (see PG&E Wants to Give Away More Money, See Fewer Email Attachments).

That's led to lots of new investment in more efficient equipment, cooling systems and power delivery systems, as well as a push into server virtualization and new sensor and control networks to keep a closer eye on data center energy use (for examples, see stories here, here, here and here).

Server makers see the data center energy crunch as an opportunity to speed up the typical cycles for replacing old equipment by offering more energy-efficient gear (see Will Energy Accelerate the Computer Refresh Cycle?).

Abbott Cuts Water Use by 1B Gallons a Year

At the end of September, we wrote about how medical supply giant Abbott cut oil and gas consumption by 35 percent compared to a 2006 baseline though, in part, waste heat technologies.

Now, the company says it is saving one billion gallons of water a year, a key milestone considering that some of its facilities are in water-starved Singapore. The company set out to get to 40 percent below its water consumption of 2004 by 2011 but it's already there. Water consumed in manufacturing processes is down 37 percent.

How did the company accomplish it? Through a variety of technologies like installing more water-efficient scrubbers for controlling dust in a Michigan facility or by installing tighter-fitting pipes in other facilities. An Arizona plant implemented leak tags to ensure that leaks are detected and fixed at a more rapid rate.

Conserving water saves energy too – around 5 percent of California's power revolves around transporting water. (The figure climbs to 19 percent if you add heating.) IBM is working on a number of water projects these days as well. Carbon, water and energy are all interrelated, according to IBM.

A Thought for Food

For every one degree Celsius increase in global temperature, there's a 10 percent decrease in crop yield.

Crop yields could be down by one-third to one-half by 2100, when the global population is likely to be considerably larger than it is now, said David Battisti, a professor of atmospheric sciences at the University of Washington. Battisti was a speaker at a geoengineering workshop at MIT Friday.

A recent study shows that U.S. crop yields are likely to decrease somewhere between 30 percebt and 82 percent by the end of the century, depending on the pace of global warming.

Battisti said that rising sea levels and increasingly destructive droughts and flooding caused by global warming aren't severe enough problems to convince him to consider drastic measures like geoengineering – deliberately altering the climate to counteract our unintended alterations. The impact of global warming on global food production, however, is another matter. "It's the one thing that scares me," he said.

There's not a lot of unexploited viable cropland left, and we already have a billion people malnourished today, he said.

While today's food security issues probably have more to do with political and economic factors affecting food distribution networks than they do with crop yields, the larger picture Battisti paints is scary. I hope a lot more research focuses on the problem. This also raises the stakes in the biofuel-versus-food debate.

The MIT workshop addressed the questions of whether geoengineering is possible and whether we should attempt it. The consensus was that precious little science has been done on geoengineering, what science is emerging is revealing that geoengineering is highly risky and uncertain, global warming is so bad that we need to consider geoengineering anyway, and we need to get busy with research on the problem. Several scientists expressed concern that we won't be able to reduce the uncertainty in the time we have left.

The issue of the geoengineering moral hazard – whether taking geoengineering seriously leads people to weaken their resolve on emissions reductions – was also discussed at the workshop (see previous post).

Eric Smalley is the editor of Energy Research News. He has written about technology since 1987 and has freelanced for many publications including Discover, Scientific American, Wired News and The Boston Globe on topics ranging from quantum cryptography to global warming.

Interim Chief for Oerlikon Solar

Jeannine Sargent, who took over the fledging solar business at Swiss thin-film factory equipment company, Oerlikon Solar, in 2007, is leaving.

The company issued a brief announcement about her departure, and the appointment of Jurg Henz as the interim CEO. Henz has been an executive within Oerlikon. 

Sargent joined Oerlikon Solar after serving as the executive vice president and general manager of Veeco Instruments' metrology and instrumentation business.

Oerlikon develops equipment for making amorphous-silicon solar panels, and has 10 customers with over 450 megawatts of generation capacities. 

The solar company saw its sales decline by 2 percent for the first nine months of this year to reach CHF 432 million.

Oerlikon won two big contracts during the second quarter: a 120-megawatt contract with Nano solar Technologies in Russia and a 30-megawatt contract with HelioSphera in Greece.

But it didn't sign any major contracts in the third quarter, and is expecting a weak forth quarter. 

For 2010, it expects to see a bit more orders but flat sales.

Having Our Coke and Drinking It Too; Carbon Capture and Sequestration

Virtually everyone reading this knows that coal is environmentally bad. It’s the dirtiest of the fossil fuels in terms of carbon emissions and historically the primary cause of acid rain, not to mention negative land and water impacts where it’s mined. But coal is also plentiful, domestically produced, and as a result, cheap. We rely on it to generate more than half of the electric power in the U.S., including the energy operating the computer on which I’m writing this blog. However, the sad but increasingly unavoidable fact is that as a power source, coal is probably here to stay for some time.

Coal will remain with us because there is simply too much global demand for electrical power to fill the gap with renewable sources such as wind, solar, hydro, and nuclear options, even assuming massive scale ups in all of these technologies. Not only the US, but China and India and other countries also rely on coal for power, with the expectation of even more use of it. According to the Natural Resources Defense Council (NRDC), globally, coal fired plants now under construction or planned during the next 20 year will, during their operational lifetimes, emit more atmospheric carbon than all the coal combusted to date.

Intriguing as coal free – alternative energy scenarios, like Ed Mazria’s solar and conservation based, 2030 Challenge are, we simply have to find a way to burn coal more cleanly to make a dent in global warming. Such clean coal processes generally fall under the rubric of Carbon Capture and Sequestration, or CSS.

In one form or another, this is a multi step process requiring: a) removal of CO2 before – or – after coal combustion, b) compression of it into a liquid, c) piping it to a safe place, where d) it can be injected/sequestered permanently, generally underground. This storage place might be an oil field, a deep saline aquifer, or in some scenarios, in the ocean depths.

CSS is not a simple process, but neither are the alternatives. Society ran out of simple, cheap energy solutions some time ago. Rather, we exploited seemingly cheap energy solutions without regard to their actual environmental costs. Respected environmental organizations such as the NRDC support the role of CCS in mitigating the worst impacts of climate change. Stanford’s Global Climate and Energy Project among other educational institutions, is actively developing CCS technologies.

In the past, there have been arguments that CCS is some sort of smoke and mirrors scheme invented by the power industry to maintain business as usual. While there may be truth to such wishful thinking by some in the power industry hoping for a silver bullet, others are moving ahead with sizeable CCS utility pilot projects; among them Southern Company.

One thing that CCS won’t be is cheap. In all likelihood it will drive up the cost of coal fired power considerably. The good news is that this makes other renewable sources such and solar and wind technologies more price competitive. But if past experience with scrubbers on power plants is any indicator, costs should come down as R & D drives innovation and scale up creates efficiencies. However, any way you look at it, renewable, fossil, or nuclear, power is going to cost a lot more more in the future, making energy efficiency and conservation all the more important.

A final note on terminology. Excuse the somewhat inaccurate reference to “coke” in the title above. The coal coking process and coal combustion to produce power are not the same, but I couldn’t resist the pun.

Eric Carlson is a greentech analyst, consultant and architect, based in Washington, D.C. and Seattle.

Odersun to Make CIS-on-Copper Goods in China

Ever since the Chinese government began discussing solar incentives earlier this year, non-Chinese companies have wondered how they could get a slice of this potentially huge market.

The consensus so far is it would be tough to crack the Chinese market because it's so new and because the government's policies are seen as measures to help its own solar manufacturers.

Teaming up with Chinese companies, then, seems a good strategy. Odersun, a German thin-film developer, is taking that route by creating a joint venture with Advanced Technology & Materials (AT&M), Odersun said Monday.

The Beijing-based joint venture would produce solar cells and panels using Odersun's technology, which deposits copper-indium-disulphide on copper stripes that are only 1 centimeter wide and 0.1 millimeter thick.

Odersun and AT&M have known each other since 2004, when AT&M invested in the German company.

The handful of non-Chinese companies that have announced plans for the Chinese market include Enfinity, a Netherlands-based project developer that has teamed up with LDK Solar and China Guandong Nuclear Power for a 10-megawatt project in Dunhuang City, Gansu province.

Two months ago, Tempe, Ariz.-based First Solar said it had a deal with the Chinese government to build a 2-gigawatt solar power plant in Inner Mongolia.