SolidNode

A diagram of the design of the CLUMEQ Colossus supercomputer, from a recent presentation by Marc Parizeau of CLUMEQ.

Here’s one of the most unusual data center designs we’ve seen. The CLUMEQ supercomputing center in Quebec has worked with Sun Microsystems to transform a huge silo into a data center. The cylindrical silo, which is 65 feet high and 36 feet wide with two-foot thick concrete walls, previously housed a Van de Graaf particle accelerator. When the accelerator was decommissioned, CLUMEQ decided to convert the facility into a high-performance computing (HPC) cluster known as Colossus.

We first noted the development of the CLUMEQ site earlier this year when Marc Hamilton of Sun discussed its unique design, but offered scant details. Additional information about the design of the facility and its cooling system were discussed at the Sun HPC Consortium last month in Portland, Oregon.

The CLUMEQ Colossus cylinder features an interior “hot core” (as opposed to a hot aisle) in the center of the building and uses the outside ring of the facility as the cold air plenum. The cabinets are arranged in a ring on each floor, facing the outside of the silo. The floors supporting each ring of cabinets are comprised of grates rather than solid flooring to facilitate airflow through the facility.

The cooling coils and air handlers are located in the basement. Chilled air flows upward through the outside cold aisle and through the racks of servers. The waste heat exits the rear of the racks into the hot core, and is returned to the basement via the cold aisle.

The air flow pattern is maintained through differential air pressure – maintaining a higher air pressure in the cold aisle than the hot aisle. This keeps the air moving through the facility, which has a blowing capacity of 180,000 CFM and can cool up to 1.5 megawatts of electrical load. Up to 300 kilowatts of cooling capacity can be supplied by free cooling using fresh air from outside the facility.

“CLUMEQ silo totally blows up the paradigm of data center design,” says Nicolas Dube of Sun, who began work on the project as a graduate student at Universite Laval in Quebec. “The silo, by itself, is the CRAC (computer room air conditioner). The whole facility cools itself.”

As for computing horsepower, Colossus will have a peak of 86 teraflops of compute power. It’s equipped with a Sun Constellation HPC systems featuring 10 fully loaded Sun Blade 6048 chassis, 1 petabyte of Lustre storage and Sun J4400 storage arrays.

The data center racks are spread over three floors, with the switches on the second floor to keep the cable runs as short as possible.

For a full description of the CLUMEQ design, check out this video from Sun, which runs about 6 minutes.

Additional details are available in PDFs of presentations by Marc Parizeau of CLUMEQ and Nicolas Dube of Sun.

OTHER STORIES YOU MAY LIKE:

• Who Has The Most Web Servers?
• The Gallery of Exploding Servers
• Inside the ‘James Bond Villain’ Data Center
• Google’s Chiller-Less Data Center
• Inside a Cloud Computing Data Center

Original post:
Wild New Design: Data Center in A Silo

Popularity: 12% [?]

The Moobull container (left) is a new modular offering for high performance computing from Bull, featuring its water-cooled Cool Cabinet Door.

There’s a new entry in the market for data center containers from Bull, the French IT conglomerate. The Mobull container is designed to house up to 1,620 servers and support power loads nearing 40kW per rack. The system, which is available in 20- or 40-foot sizes, employs Bull’s water-cooled  Cool Door technology on the rear door of server cabinets. For more ifnromation, see Bull’s announcement.

Original post:
Bull Debuts Mobull Container for HPC

Popularity: 20% [?]

A look at the Aurora supercomputer from Eurotech, which uses liquid cooling and SSD memory.

There was lots of news last week from the SC09 conference in Portland. Here’s a roundup of some items that we didn’t get a chance to cover last week:

• Eurotech, which previously focused on wearable computers and other “special purpose platforms,” has entered the HPC/supercomputing fray with Aurora, a design that uses direct liquid cooling and solid state storage to offer energy-efficient horsepower (see photo above).
• The new product generating the most buzz and discussion was the rollout of its new Fermi GPU product for high performance computing from NVIDIA, whose presentations on the technology drew large crowds.
• Former vice president and Nobel Prize winner Al Gore was the keynote speaker, and reminded attendees that he attended the first SC event in 1988. Gore’s message was that it was “up to HPC experts to make global warming real to people, as well as do the modeling for climate change,” The Register noted in its coverage. At InsideHPC, John West said Gore’s speech was “good for the community — like him or hate him, SC09 got loads of international press coverage that we would have never gotten if Al Gore hadn’t spoken at our conference this year.”
• While the Top 500 ranks the most powerful supercomputers, the Green 500 ranks supercomputers based on energy efficiency. At SC09 it was announced that the Green 500 is expanding the definition of a supercomputer to include a wider spectrum of the high-end computing world with the “Little Green500″ list and opening its doors to innovation with two new exploratory lists: the “Open Green500″ and “HPCC Green500.”
• Fusion-io announced that two unnamed government organizations will each deploy hundreds of terabytes of the company’s solid-state storage capacity, which in these configurations is capable of sustaining over one terabyte per second (1TB/s) of aggregate bandwidth with access latencies under 50 microseconds.

Original post:
SC09 Wrap: Aurora, Gore, Fusion i/o, Green 500

Popularity: unranked [?]

The twice-a-year list of the Top 500 supercomputers documents the most powerful systems on the planet. Many of these supercomputers are striking not just for their processing power, but for their design and appearance as well. Here’s a look at the top 5 supercomputers, their specs, and some cool photos.

Original post:
Fast and Fierce: 5 Awesome Supercomputers

Popularity: unranked [?]

What’s it like to perform a processor upgrade on one of the world’s most powerful supercomputers? In this video, Al Linger from Cray, Inc. walks through the upgrade process of the Kraken supercomputer at the Oak Ridge National Laboratory, which holds the number three spot in the new Top 500 list of the world’s most powerful supercomputers. Linger is upgrading Kraken’s quad-core AMD Opteron chips to Opterons with six cores. This video runs about 4 minutes.

For more coverage of information about supercomputing, check out our High Performance Computing Channel. For additional video, check out our DCK video archive and the Data Center Videos channel on YouTube.

Original post:
Upgrading A Supercomputer

Popularity: 8% [?]

The new Red Sky supercomputer as Sandia National Laboratories just debuted as the 10th fastest supercomputer on the Top500 list, with a sustained performance of 429.9 teraflops. Red Sky consists of 68 cabinets of Sun Constellation gear, with up to 96 nodes and 678 cores per rack. Each cabinet  can each require up to 32 kilowatts of energy at full load.

But the system is notable not just for its power, but for its energy efficiency. Red Sky has an estimated Power Usage Effectiveness (PUE) of 1.035. That’s not a typo - a claimed PUE of 1.035. How is this possible? Red Sky uses the Sun Cooling Door (also known as Project Glacier) designed jointly by Sun Microsystems and Emerson Network Power, which was demonstrated at the SC08 show. The Sun Cooling Door 5600 attaches to the back of cabinets and uses an inert refrigerant gas called R134. The unit is supported by a Liebert XD pumping unit. The passive design that doesn’t require additional fans to circulate air, saving on energy used to power the fans.

Here’s a time-lapse video of the assembly of Red Sky (link via Marc Hamilton). This video runs about 5 minutes.

For more coverage of information about supercomputing, check out our High Performance Computing Channel. For additional video, check out our DCK video archive and the Data Center Videos channel on YouTube.

Original post:
Red Sky: Supercomputing and Efficiency Meet

Popularity: unranked [?]

Here’s a roundup of news announcements from the data center and hosting industry:

• Key milestone to 100 Gigabit Ethernet demonstrated. At the SC09 conference in Oregon Monday Infinera, Internet2, Juniper Networks and Level 3 demonstrated 100 Gbps data transport between Seattle and the SC09 show floor in Portland.  The 100 Gbps of test data was sent via a single slot on the Juniper T1600 Series Core Router, populated with a new 10
  

  Popularity: 6% [?]

HP this week launched several new products for high performance computing at the SC09 conference, including servers, storage, software and network products.A new ProLiant BL2

Popularity: unranked [?]

The Jaguar supercomputer at Oak Ridge National Labs, which is the new leader on the Top 500 list of the most powerful supercomputers.

The new Top 500 list of the most powerful supercomputers has been released in conjunction with this week’s SC09 conference, and there’s a change atop the leaderboard. After two years as runner-up, the Jaguar supercomputer located at the Department of Energy’s Oak Ridge Leadership Computing Facility has overtaken the RoadRunner system at Los Alamos labs to claim the top spot.

Jaguar, which was upgraded earlier this year, posted a 1.75 petaflop/s performance speed running the Linpack benchmark. One petaflop/s refers to one quadrillion calculations per second. The Roadrunner system at Los Alamos was the world’s first petaflop supercomputer, but its performance of 1.04 petaflops was down slightly from 1.105 petaflops in June 2009 due to a repartitioning of the system.

See the new Top 500 list for additional details.

Original post:
Jaguar Bumps RoadRunner in Top 500

Popularity: unranked [?]

Indiana University has opened its new 82,700 gross square foot data center, which includes three 11,000-square-foot computer equipment rooms, 13 miles of cabling, two flywheels and two diesel generators, and is engineered to withstand an F5 tornado. The facility, which cost $32.7 million to build, will house the university’s high-performance computing operation and supercomputers, known as Big Red and Quarry. “IU has a real data center now,” says Matthew Link, director of systems at the university, who said the expansion was critical to continued growth of the school’s IT systems. During the moving-in process, the IU team moved 41 racks of equipment with more than 7,000 computing cores in seven hours. This video runs about 5 minutes.

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

Original post:
Indiana’s New Supercomputing Center

Popularity: 6% [?]