October 31, 2012
Air-cooled servers may soon go the way of the single-core CPU. In high performance computing datacenters, the hottest new trend in energy efficiency is warm water cooling. IBM, Eurotech, and a handful of other vendors have paved the way with this technology and now Appro has announced its own solution in an attempt to set itself apart from the competition.
Warm water cooling is benefiting from a confluence of industry trends that have raised the profile of the technology. Especially for the hotter, denser HPC systems that are being shoehorned into datacenters these days, warm water technology can offer an optimal solution, balancing somewhat higher up-front cost with a much lower lifecycle cost.
Compared to traditionally air cooled systems, liquid cooling (at any temperature) offers better energy efficiency, thus lowering the power bill. That's because water has much better thermal properties than air, requiring a lot less of it to cool a given piece of hardware. Using warm water instead of cool water has the additional advantage of doing away with a water chiller unit, greatly simplying the plumbing, not to mention reducing installation costs.
While warm water doesn't have the chilling capacity of cold water, as long as you get liquid in close proximity to the hardware, it can cool even the hottest processors. Even water as warm as 45C (113F) can effectively cool a modern processor. And like cool water setups, the warmer outlet water from the servers can be reused to heat the datacenter and surrounding facilities.
The other development that is lighting a fire under this technology is the proliferation of high-wattage chips. The continued demand for performance means server chips are continuing to push the power envelope. Fast, high performance x86 CPUs can easily reach 130 watts. In a dense two-socket (or worse, four-socket) system, heat can build up quickly – all the more so when you consider more memory chips are needed to feed the growing number of cores on these processors.
In the HPC realm, an additional burden has been added with the advent of accelerators: GPUs and soon the Intel Xeon Phi. Although the chips themselves aren't much hotter than a top-bin CPU, the inclusion of multiple gigabytes of graphics memory on an accelerator card pushes these devices well past the 200 watt realm. Once you start gluing a couple of these together on the same motherboard, along with their CPU hosts and main memory, all of a sudden you have over a kilowatt of hardware in a very small space.
It is in this environment that Appro has decided to offer its warm water cooling option, which it has dubbed Xtreme-Cool. The company claims it will reduce energy consumption by 50 percent and provide a PUE below 1.1. According the Appro, their cooling gear is designed to fold seamlessly into the company's Xtreme-X blade system, which previously was offered only with standard air cooling or chilled water setups. But the Xtreme-Cool design is such that it will actually fit into any standard computer rack.

It's especially geared for the dense blade designs available in Xtreme-X, the Appro platform that features dual-processor nodes, 80 of which can be fit into a standard 42U rack. Since the company will be offering a Xeon Phi option for these blades when the chips become available, there will be an extra incentive for customers to consider the warm water option. GPU accelerators will be supported as well.
Xtreme-Cool is different from most of the warm water cooling solutions out there, inasmuch as it's built for standard rack enclosures. The heart of the system is the RackCDU (rack cooling device unit), a radiator-like component that sits on the inside of the rack enclosure. Two sets of tubes run from the unit to the server blade. One set feeds the warm water to the servers; the other transfers the server-heated water back to the RackCDU for cooling to ambient temperatures.
The tubes that go into the server wrap around the cold plates on top of the processor and memory components, the primary sources of heat on the motherboard. Dripless interconnectors are used for reliability. Appro does the entire installation, so from the facility manager's point of view, it's plug and play.
As mentioned before, Xtreme-Cool can be adapted to standard, non-Appro racks. Most other solutions, like that of IBM's and Eurotech's, are custom designs, architected to fit their particular blade systems. For example, the IBM solution, which is being used in SuperMUC, a three-petaflop supercomputer cluster constructed from iDataPlex servers, uses the company's own hot water cooling system.
That one consists of custom-fitted aluminum plates that lay over the server motherboard. This design actually does a somewhat better job at extracting the heat – Appro says their solution will only extract about 80 percent of it – but at a cost that is considerably higher.
Appro is trying to hit the sweet spot here, designing a system that does a good job at heat extraction, but at a price point that they believe will deliver a faster return on investment than more custom designs. In truth, the company has not specified the price premium on the Xtreme-Cool option yet, but according to Appro marketing director Maria McLaughlin it will be "much cheaper than the proprietary systems."
Xtreme-Cool systems won't start shipping until the first quarter of 2013, but Appro will be demonstrating the product at the Supercomputing Conference (SC12), on November 12 to 15.
Jun 18, 2013 |
The world's largest supercomputers, like Tianhe-2, are great at traditional, compute-intensive HPC workloads, such as simulating atomic decay or modeling tornados. But data-intensive applications--such as mining big data sets for connections--is a different sort of workload, and runs best on a different sort of computer.
Read more...
Jun 18, 2013 |
Researchers are finding innovative uses for Gordon, the 285 teraflop supercomputer housed at the San Diego Supercomputer Center (SDSC) that has a unique Flash-based storage system. Since going online, researchers have put the incredibly fast I/O to use on a wide variety of workloads, ranging from chemistry to political science.
Read more...
Jun 17, 2013 |
The advent of low-power mobile processors and cloud delivery models is changing the economics of computing. But just as an economy car is good at different things than a full size truck, an HPC workload still has certain computing demands that neither the fastest smartphone nor the most elastic cloud cluster can fulfill.
Read more...
Jun 14, 2013 |
For all the progress we've made in IT over the last 50 years, there's one area of life that has steadfastly eluded the grasp of computers: understanding human language. Now, researchers at the Texas Advanced Computing Center (TACC) are utilizing a Hadoop cluster on its Longhorn supercomputer to move the state of the art of language processing a little bit further.
Read more...
Jun 13, 2013 |
Titan, the Cray XK7 at the Oak Ridge National Lab that debuted last fall as the fastest supercomputer in the world with 17.59 petaflops of sustained computing power, will rely on its previous LINPACK test for the upcoming edition of the Top 500 list.
Read more...
05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.
04/15/2013 | Bull | “50% of HPC users say their largest jobs scale to 120 cores or less.” How about yours? Are your codes ready to take advantage of today’s and tomorrow’s ultra-parallel HPC systems? Download this White Paper by Analysts Intersect360 Research to see what Bull and Intel’s Center for Excellence in Parallel Programming can do for your codes.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?
Join our webinar to learn how IT managers can migrate to a more resilient, flexible and scalable solution that grows with the data center. Mellanox VMS is future-proof, efficient and brings significant CAPEX and OPEX savings. The VMS is available today.