YGOPRO 2 AI SERVER FULL
Dual GPU baseboard with 16 GPU fully connected at full NVLink speed of 300GB/sįigure 2 shows the physical implementation. This topology enables all 16 V100 Tensor Core GPUs to be fully connected every GPU can communicate with any other GPU simultaneously at the full NVLink speed of 300GB/s. Each HGX-2 server platform consists of a pair of baseboards with full-speed connectivity offered via 48 NVLink ports between the two baseboards.įigure 1 shows the logical diagram of the dual GPU baseboard sub-system. (More on the Volta GPU architecture can be found here).
Each V100 Tensor Core GPU includes six NVLinks each NVLink routes to one of the six NVSwitches. Each NVSwitch is a fully non-blocking NVLink switch with 18 ports so any port can communicate with any other port at full NVLink speed. The baseboard hosts eight V100 32GB Tensor Core GPUs and six NVSwitches. The GPU baseboard represents the key building block to the HGX-2 server platform. HGX-2 Server Architecture The GPU baseboard We’ll also discuss how we cooperate with ecosystem partners in bringing their HGX-2 based servers to the cloud, walk through the benefits to end-user applications, and discuss platform optimizations targeted for cloud data center deployment. Let’s dig deeper into the HGX-2 server architecture.
HGX-2 comes with multi-precision computing capabilities, allowing high-precision calculations using FP64 and FP32 for scientific computing and simulations, while also enabling FP16 and INT8 for AI training and inference. NVIDIA is introducing HGX-2, our latest, most powerful cloud server platform for AI and High-Performance Computing (HPC) workloads, to meet these new requirements. The continuing explosive growth of AI model size and complexity means the appetite for more powerful compute solutions continues to accelerate rapidly.