German University Gets SiCortex HPC To Power Research
- By Dian Schaffhauser
The Otto-von-Guericke University of Magdeburg in Germany will be implementing a SiCortex SC5832 high productivity computer to further advance the university's scientific programs in a number of key areas. Scientists will use the SC5832 to assist in the calculation of quantum spin systems, a series of projects that have been accumulating data for 20 years.
In addition, specific applications will run on the SC5832 in the areas of many-body physics and materials science, including the study of nano- and microstructures in semiconductor materials, random number generation, and eutectic crystallization. The university's Laboratory of Fluid Dynamics and Technical Flows will use the computer to advance research in the field of fluid dynamics.
The university considered systems from six computer manufacturers but chose the SiCortex hardware for its performance, scalability, and energy efficiency. Qualifying tests run prior to purchase demonstrated that its overall system bandwidth was 18 times greater than competing systems for their particular applications.
In terms of energy efficiency, said Rolf Knocke, director of the computing center, "We capped power consumption at 60 kilowatts for the bid, and that limit was a problem for many of the proposed systems.... Not only did the SiCortex system meet our power requirements, it came in 40 kilowatts under the limit--quite impressive considering the performance levels of the machine. The expected cost savings over time by running this energy efficient system was also a factor in our decision. Overall, we're very optimistic about how the computer will perform, and are proud to be the first university in Europe to install this dedicated HPC machine."
The university will be buying the computer from SiCortex partner Megware Computer GmbH.
Dian Schaffhauser is a senior contributing editor for 1105 Media's education publications THE Journal and Campus Technology. She can be reached at firstname.lastname@example.org or on Twitter @schaffhauser.