Software stack drives energy-efficient high-performance computing

Marvell now offers availability of NVIDIA GPU support on its ThunderX family of Arm-based server processors. Following the company's June announcement to bring CUDA to the Arm architecture, they has collaborated with NVIDIA to port its CUDA-X AI and HPC libraries, GPU-accelerated AI frameworks and software development tools to the ThunderX platform. The computational performance and memory bandwidth of ThunderX2, the company's newest 64-bit Armv8-A based server processor, coupled with the parallel processing abilities of NVIDIA GPUs offer a compelling path to energy-efficient exascale computing.

AI and ML continue to become necessary technology components to data centre server demands at the cloud and network edge. To answer these evolving AI and ML workloads, as well as the most challenging and complex problems in science and research, supercomputers require processors that are optimised to offer cutting-edge throughput, application latency and power.

With a primary focus on computational science applications comprising GROMACS, NAMD, MILC and LAMMPS, the ThunderX2 configurations are showing compelling performance with an enhanced capability to drive higher and further efficiency coupled application results in a GPU-enabled system.

“NVIDIA GPU support for our ThunderX2 server processor brings clear, differentiated value to meet the distinctive performance and power requirements of the exascale computing era,” said Gopal Hegde, vice president and general manager, Server Processor Business Unit at Marvell Semiconductor, Inc. “The availability of NVIDIA GPU acceleration on the Arm architecture will further expand the ThunderX2 ecosystem for HPC, cloud computing and edge markets, spurring innovation across low-level firmware through system software to commercial ISV applications.”

“The availability of CUDA acceleration for ThunderX2 processors marks a significant milestone in bringing the power efficiency and high performance of the Arm architecture to the infrastructure market,” said Chris Bergey, senior vice president and general manager, Infrastructure Line of Business at Arm. “The breadth and depth of innovation across the ecosystem enables an easy migration path and robust support for existing and future GPU workloads from the edge to the cloud.”

“NVIDIA GPU-accelerated computing on Arm provides customers worldwide with greater choice in building next-gen AI-enabled supercomputers,” said Ian Buck, general manager and vice president of Accelerated Computing at NVIDIA. “Combining NVIDIA’s unmatched platform for AI and HPC with Marvell’s powerful ThunderX2 Arm-based server processors is already delivering impressive application performance.