Enterprise computing, storage, networking solutions provider, Supermicro announced its new systems. The new systems are based on NVIDIA Ampere architecture GPUs and 3rd Generation Intel Xeon Scalable processors with built-in AI accelerators. New solutions are ideal for AI applications that require low latency and high performance.
Reduced energy usage
The 2U NVIDIA HGX A100 4-GPU system is designed for modern AI training clusters at scale with high-speed CPU-GPU and GPU-GPU interconnect. Supermicro 2U 2-Node system shares power supplies and cooling fans which allows organizations to reduce energy usage, reduce carbon emissions. It also supports a range of discrete GPU accelerators, which can be matched.
The 2U NVIDIA HGX A100 server is based on the 3rd Gen Intel Xeon Scalable processors. It features Intel Deep Learning Boost technology and is designed for analytics, training, and inference workloads. It is capable of delivering up to 2.5 petaflops of AI performance. It comes with four A100 GPUs fully interconnected with NVIDIA NVLink that provides up to 320 GB of GPU memory.
Intel Optane Persistent Memory
Super Micro also stated that these systems are ideal for high-performance clusters with advanced thermal and cooling designs. Liquid cooling and Intel Optane Persistent Memory are also supported on this platform, allowing larger models to be held in memory, close to the CPU, before processing on the GPUs. Charles Liang, president and CEO of Supermicro said,
“Supermicro engineers have created another extensive portfolio of high-performance GPU-based systems that reduce costs, space, and power consumption compared to other designs in the market. With our innovative design, we can offer customers NVIDIA HGX A100 (code name Redstone) 4-GPU accelerators for AI and HPC workloads in dense 2U form factors. Also, our 2U 2-Node system is uniquely designed to share power and cooling components which reduce OPEX and the impact on the environment.”