In the first episode of Supermicro Tech Talks, they take a detailed look at their 4U System featuring NVIDIA HGX A100 8-GPU. This system is ideal for large-scale deep learning, training, and neural network model applications, and with the newest version of NVIDIA® NVLink™ and NVIDIA NVSwitch™ technologies, it can deliver up to 5 PetaFLOPS of AI performance in a single 4U server. For the most demanding AI workloads, Supermicro builds the highest-performance, fastest-to-market servers based on NVIDIA A100™ Tensor Core GPUs.
Supermicro's GPU based systems are purpose built enterpise servers that are optimised for rack density, GPU connectivty and the challenging power and thermal demands of this system. The primary benefits of this server are the optimised density of the 2U form factor, the NVIDIA A100 HGX baseboard which has 4 A100 GPUs on them with NVLink™ interconnect and 2 x AMD EPYC™ CPUs with PCI-E Gen 4.0.
Watch the full episode below:
Browse our Supermicro A100 servers here.