Cray Sets New Supercomputing Record with HLRS and Ansys

Hazel Hen, the new CRAY XC40-System of the High Performance Computing Center Stuttgart (HLRS), delivers a peak performance of 7.42 Petaflops (quadrillion floating point operations per second).

Hazel Hen, the CRAY XC40-System of the High Performance Computing Center Stuttgart (HLRS), delivers a peak performance of 7.42 Petaflops (quadrillion floating point operations per second).

ANSYS, HLRS and Cray have pushed the boundaries of supercomputing by achieving a new supercomputing milestone by scaling ANSYS software to 172,032 cores on the Cray XC40 supercomputer, hosted at HLRS, running at 82 percent efficiency. This is nearly a 5x increase over the record set two years ago when Fluent was scaled to 36,000 cores.

Our high-performance computing technology partnership with HRLS is delivering cutting-edge simulation capabilities,” said Wim Slagter, director of HPC and cloud marketing at ANSYS. “With the state-of-the-art resources and support from Cray, as well as access to government, industry and academia, we can use HPC to solve even more complex and challenging problems across any industry.”

By scaling ANSYS Fluent to over 172,000 computer cores on their Hazel Hen a Cray XC40 supercomputer, HLRS is enabling organizations to create innovative and groundbreaking complete virtual prototypes of their products faster and more efficiently than ever.

By leveraging HPC, companies can rapidly iterate their products. Even though most organizations do not have access to this extreme core count level yet, users across all computing platforms from HPC clusters over Cloud to engineering desktops can take advantage of the breakthroughs that speed up computing at all levels.

Since announcing their partnership in 2015, ANSYS has worked with HLRS and Cray to profile and benchmark ANSYS simulation software for extreme HPC scalability and capability. The partnership not only ensures that ANSYS simulation software scales to extreme loads, but also broadens the scope of simulations, allowing for applicability to a much broader set of real-world problems and products.

We see the role of HLRS as vital for industrial innovation,” said Michael M. Resch, HLRS director. “We not only provide the HPC platforms for industrial companies and scientific organizations 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.”

The new record is critical as the demand for HPC to solve large-scale simulation challenges is growing across industries, especially in the aerospace and automotive industries, where product simulation models are becoming larger and more complex. We can now tackle more complex and broader systems level simulation needed to develop the smarter and greener products of tomorrow.

This record-setting scaling of ANSYS software on the Cray XC40 supercomputer at HLRS proves that close collaborations with customers and partners can produce exceptional results for running complex simulations,” said Fred Kohout, senior vice president and chief marketing officer at Cray. “The highly-integrated architecture of the Cray XC40 and its Aries interconnect are designed for applications at any scale, and allow scientists and engineers to push the boundaries of advanced simulations.”

Sign up for our insideHPC Newsletter