Today Supermicro announced the company is among the first to adopt the NVIDIA HGX-2 cloud server platform to develop the world’s most powerful systems for artificial intelligence and high-performance computing.
To help address the rapidly expanding size of AI models that sometimes require weeks to train, Supermicro is developing cloud servers based on the HGX-2 platform that will deliver more than double the performance,” said Charles Liang, president and CEO of Supermicro. “The HGX-2 system will enable efficient training of complex models. It combines 16 Tesla V100 32GB SXM3 GPUs connected via NVLink and NVSwitch to work as a unified 2 PetaFlop accelerator with half a terabyte of aggregate memory to deliver unmatched compute power.”
From natural speech by computers to autonomous vehicles, rapid progress in AI has transformed entire industries. To enable these capabilities, AI models are exploding in size. HPC applications are similarly growing in complexity as they unlock new scientific insights. Supermicro’s HGX-2 based systems will provide a superset design for datacenters accelerating AI and HPC in the cloud. With fine-tuned optimizations, Supermicro’s HGX-2 server will deliver the highest compute performance and memory for rapid model training.
As AI model complexity and size are exploding, researchers and data scientists need new
levels of GPU-accelerated computing,” said Ian Buck, vice president and general manager of
accelerated computing at NVIDIA. “HGX-2 provides the power to handle these massive new
models for faster training of advanced AI, while saving significant cost, space and energy in the
datacenter.”
