Intel and HP Team up to Breed Warm Water Xeons

These little fish know how to stay cool…

Intel - logo A new HPC data center is going in at the National Renewable Energy Lab featuring two types of Xeons and warm water cooling courtesy of HP. Intel’s Xeon E5 series of server processors will be at the heart of this new cluster that is slated to reach its peak compute capacity in mid-2013. Additionally, 600 of Intel’s Xeon Phi or “Knights Corner” co-processors will be there to aid the approximately 3200 Xeon CPUs to create the world largest supercomputer dedicated to renewable energy research.

The really interesting part of this Petaflop capable installation is the new warm water cooling system that HP designed for it. While Intel was short on specific details about this cooling solution they did mention that the waste heat from the data center will be used to heat the rest of the building that the data center is located in. The one point that Intel did trumpet about this cooling solution is its efficiency, which is significantly below average in terms of additional energy used just for cooling.Intel Xeon Phi Logo

Intel is pushing its yet to launch Xeon Phi coprocessors just about any way it can these days, and this installation is no exception. In this case the researchers at the NREL were able to port a half-million lines of code to Intel’s MIC architecture in just a few days.

The drum beat just keeps on going as Intel closes in on HPC market. With its E5 series of processors holding the line in the CPU market, and its Xeon Phi coprocessors are probing GPU space. If one thing is clear, it’s that Nvidia’s K20 is going to have to be something really special to keep its Tesla line competitive with Intel’s Xeon Phi.S|A

The following two tabs change content below.
Thomas Ryan is a freelance technology writer and photographer from Seattle, living in Austin. You can also find his work on SemiAccurate and PCWorld. He has a BA in Geography from the University of Washington with a minor in Urban Design and Planning and specializes in geospatial data science. If you have a hardware performance question or an interesting data set Thomas has you covered.