Paul Brenner from the University of Notre Dame Center for Research Computing has developed another novel approach to recycling waste heat. In a recent presentation at an open source conference in Oakland, Brenner said he had placed a rack of high-performance computing (HPC) nodes at a local municipal greenhouse, the South Bend Greenhouse and Botanical Garden, to help heat the flowers and plants in the facility. Here's a description from Sun's Mike Stevens:

Based on early prototype work which involves placing single rack in the greenhouse, the idea looks like a promising way to reduce natural gas heating requirements for the facility. Brenner has shown he can use grid scheduling software to deliver a desired temperature (within a range, of course) by simply adding or throttling compute jobs on the greenhouse cluster, which communicates with Notre Dame via a wide-area wireless broadband connection. He has looked at humidity issues and so far they don't seem to be a problem given the ranges supported by typical compute gear. And he points out that while the greenhouse environment does not offer the highly filtered environment of a controlled datacenter, the particulate tolerance for typical compute gear is far in excess of EPA guidelines for people. Phase II will involve placing three full racks of gear at the greenhouse to significantly reduce heating costs. Notre Dame will pay the electrical costs and use the compute resources. The city saves money on heating.

Modular expansion isn't just for servers or compute capacity. Several vendors have developed products that allow data center operators to easily expand the power capacity of existing facilities, or an instant power infrastructure for container-based expansion solutions. Active Power, Inc. (ACPW) has developed PowerHouse (pictured above) which packages a flywheel UPS, switchgear, a diesel generator and fuel tanks in a shipping container.

PowerHouse was designed to provide a turnkey power and cooling infrastructure to accompany the Sun Modular Data Center S20, easily sold two containers to Tesco PLC to boost power capacity in existing data center. The PowerHouse gear is stored in a 40 foot shipping container, similar to those now being used in mobile data centers from Microsoft, Rackable, Verari, IBM and Dell.

Last summer we first began to note is the emergence of products customized for the "Blackbox economy" of container data centers. And with a growing number of facilities-based data centers running out of power capacity, modular power expansion products have application well beyond the container sector.

"There is no more generation and distribution capacity coming into New York," Eaton's Kenneth Uhlman said in March at the DataCenterDynamics New York conference. "You guys are locked. The electrical grid congestion is a major issue. You have some huge opportunities for on-site power and cooling."

Uhlman said Eaton's prefab power expansion products can be dropped on a cement pad, with six months between order and delivery. Customers can buy or lease units, and Eaton offers financing options.

"I know of no other way to incrementally increased power in a cost effective way," said Uhlman. "It gives you a good way to generate electricity without tying up the capital."

For additional video, check out our DCK video archive and the Data Center Videos channel on YouTube.

ICE, a leading electronic energy marketplace, consolidated its primary data center operations in Chicago last year from existing data centers in London and Atlanta. ICE's data centers support electronic markets for ICE Futures, the New York Board of Trade (NYBOT) and ICE's global over-the-counter markets.

Intercontinental Exchange markets offer access to a range of contracts based on crude oil and refined products, natural gas, power and emissions, as well as agricultural commodities including cocoa, coffee, cotton, ethanol, orange juice, wood pulp and sugar, in addition to currency and index futures and options.

Each PowerHouse container system includes one CleanSource UPS 1500iC system, switchgear, a 1900 kVA diesel generator and fuel tanks. The containers will be positioned beside Tesco's existing data center facility. Both containers will be manufactured offsite, which will ensure minimal disruption at the Tesco facility and pose less risk to day-to-day business operations.

Active Power’s flywheel UPS systems lend themselves very well to containerization, as the system only consumes about one quarter the footprint as compared to a conventional battery based system and can operate in an environment with wide temperatures swings (0 to 40 degrees Celsius.

Active Power has shipped more than 1,700 flywheels in UPS systems since it was founded in 1992.

So this week, Dell has taken the wraps off one of its custom server designs for cloud and hyper scale computing, known as the XS23 Cloud Server.

Here's an overview from Todd Brannon of Dell Data Center Solutions: "This product was designed for a customer that needed maximum compute density, a healthy amount of local disk and, of course, lowest power draw possible," Brannon writes. "Our architecture team threw all that in the blender and out came a 2U standard rack mount chassis that houses four dual-socket servers and twelve 3.5” hot plug drives."

Density of this type is certainly not unheard of (half depth or twin 1U’s), but by going to a 2U chassis we were able to fit it with larger, more efficient fans and stack 3 rows of full 3.5” drives across the front. So, even with a 25% higher density than general purpose blades, it provides three local spindles of 3.5” SAS/SATA disk to each server. Of course there are tradeoffs. This was expressly designed for an environment with high node failure tolerance - a cloud application. By designing out a lot of the capabilities that weren’t required (like redundant power) we were able to deliver the performance and power profile required. Efficiencies are gained by shared resources - as seen in a lot of general purpose designs available today.

In this video, Cisco's Bob Browning provides a tour of Cisco's Network Emergency Response Vehicle (NERV) and talks about how the NERV was used in the Harris Fire.

For more information, see our past coverage of Cisco. For additional video, check out our DCK video archive, the Data Center Videos channel on YouTube.

• InformationWeek notes that there will almost certainly be significant data center consolidation. In 2006, HP announced one of the most ambitious projects yet, consolidating 85 data centers worldwide into six larger centers located in Atlanta, Houston and Austin.
• Nick Carr says "cloud computing promises to turn many traditional systems-outsourcing businesses into pure commodity businesses - undifferentiated utility services."
• Drue Reeves at The Burton Group reflects on the competitive implications for Dell. "From a services perspective, the move by HP leaves Dell with few options to get into the services race. About the only possible acquisitions left are Unisys (already a Dell services partner), Accenture, Computer Science Corp, and Perot Systems. Accenture is probably the cream of the crop, but their price – thanks to HP’s move – may be more than Dell wants to pay at this point."

See additional analysis at Reuters and Silicon Alley Insider.

• A number of prominent bloggers have advanced strategies for decentralizing Twitter. I don't recall Twitter asking to be decentralized, but the service's recent performance problems have invited speculation about alternatives. Against that backdrop, Alex Iskold at ReadWriteWeb takes a look at the architecture challenge faced by twitter and similar light streaming services, and whether the cloud can improve upon relational database approaches.
• Emerson Process Management has announced plans to build a $12 million data center in Marshalltown, Iowa. The center will back up information from Emerson operations across the world, according to the Cedar Valley Courier.
• Premier Technical Services of Luray, Virginia will open a data center in Page County, creating 96 new jobs. The $16.5 million data center that will incorporate "green design," and the company plans to apply for LEED (Leadership in Energy and Environmental Design) certification from the U.S. Green Building Council for the new facility.

The data center business is doing its part to contribute to the growth of online video, as can be seen at our DCK video archive, the Data Center Videos channel on YouTube, and also at Data Center Journal's new DataCenter.tv.

So today I am adjusting to life as a one armed blogger, at least for the next few weeks. The good news is that I'm right-handed, and thus can still take notes. I've also gotten reasonably handy with Dragon Naturally Speaking dictation software, which I began using last year during a carpal tunnel flare-up. Nonetheless, posting may be slightly lighter than usual in the next several days as I get acclimated.

That decision appears to have a good one, as the new UFO files website has been widely linked in news coverage around the Web today, including a prominent link on Slashdot. In case you were wondering, thousands of files from the US government's UFO research effort, Project Blue Book, are already online. The archive includes nearly 15,000 individual UFO reports.

With the new schedule, our employees work 12 hours each day in a two-week rotation. In one week they’ll work three days, followed by four days the following week. These schedules actually provide more people on each shift throughout the entire day, and also eliminate wasted time overlaps between shifts and days with double staff. Every technician affected by the schedule change alternates three- and four-day weekends. Because the shifts are longer, each technician comes into the office seven out of every 14 days, compared with 10 out of every 14 days with the old schedules. The new schedule also streamlines our support communications, especially with regard to ownership and handoffs. We’re eliminating a few shift changes in the middle of the day, so projects and tickets aren’t as likely to bounce between shifts.

The data center in Slough, Berkshire is part of a $400 million global investment by Savvis in new facilities and network assets across Europe, Asia and North America over a two year period. The London area facility is scheduled to open in the fourth quarter of 2008.

Savvis said it decided to locate the data center outside the M25 (the London "beltway") and closer to the grid to "overcome space and power challenges often associated with central London facilities." The Slough facility will have direct connectivity to SAVVIS’ data centers near Reading and in London’s Docklands, allowing customers in those facilities to store data for disaster recovery. Slough is also the site of data centers for Rackspace and Equinix (EQIX).

"Over the past five years, SAVVIS’ European hosting business, predominately managed hosting, has grown at a compound annual rate approaching 60 percent," said Richard Warley, International Managing Director for SAVVIS. "This new data centre will provide sufficient capacity to accommodate the growth requirements of existing and new customers over the next five years. The facility will be an attractive option for those companies who currently require colocation services, but are considering managed hosting and networking solutions with SAVVIS as part of their strategic IT plans."