A new world record in supercomputing capacity broadens simulation possibilities. How’s it helping everyday engineers?
The HLRS’ Cray XC40, nicknamed Hazel Hen.
High performance computing (HPC) is quickly becoming a major part of engineering and design infrastructure. From product simulations to fundamental materials research, engineers are employing supercomputing resources to solve some of the most critical design challenges of the day.
But where is the state of the art in HPC?
Recently, ANSYS, Cray, and the High Performance Computing Center Stuttgart (HLRS) announced a new world record in HPC scaling, harnessing the power of the 172,032 cores built into the center’s Cray XC40 supercomputer. Running at 82 percent efficiency, the HLRS’ current supercomputing configuration represents a five-fold increase in performance, leaving its previous incarnation (36,000 interconnected cores) looking feeble.
But what’s this advance in supercomputing got to do with everyday engineering?
Well, whether you’re running simulation on a cloud-based HPC configuration, on the cores housed in a local workstation or on an onsite supercomputer, the ability to program a system to thread large-scale problems through available resources is critical to running efficient simulations and broadening the scope of problems that can be tackled by computers.
Take for example the innovations that are being demanded of the aerospace and automotive industries. Whether it be mandated by government or driven by consumer demand, both industries are being pushed to increase efficiency and lower environmental impact of their products. These demands are forcing companies to look to radical new designs to meet these goals. Without robust and novel simulation technologies backed by sophisticated HPC infrastructures, the development and iteration of these new design paradigms wouldn’t be possible.
“We see the role of the HLRS as vital for industrial innovation,” said Michael M. Resch, HLRS director. “We not only provide the HPC platforms for industrial companies, but also support them in developing solutions for their research and their business based on extreme HPC capabilities. This partnership is a prime example that supercomputing can be brought to bear on some of the most pressing technical challenges of the day.”