Navion 6U with NVLink – HGX H200 8 GPU
8-GPU server with NVIDIA H200 with NVLink GPUs – HGX H200™ 8 GPU
Microway’s Navion™ 8 GPU server with NVIDIA® NVLink™ Technology provides eight NVIDIA H200 Tensor Core GPUs and two AMD EPYC Processors within a 6U rackmount server platform. This system is based on the NVIDIA HGX H200 8 GPU platform.
Key Features
Built for Training Workloads
Navion 6U with NVLink is optimized for training workloads. NVIDIA NVLink technology interconnects all 8 GPUs into a single memory space. The high-bandwidth, unified memory architecture substantially benefits training workloads.
Arrives Fully Integrated with NVIDIA Software
Arrives fully ready to train with NVIDIA CUDA® SDK Optional NVIDIA AI Enterprise and NVIDIA NGC Software integration is available. Your system runs applications out of the box.
Fully Configurable NVIDIA HGX Solution
Built upon the NVIDIA HGX platform, this system is a fully configurable 8-GPU server with NVIDIA NVLink technology. The underling GPU hardware architecture is the same as used in NVIDIA DGX™ H200.
- 8 NVIDIA H200 Tensor Core GPUs with: 191GB HBM3e memory, 4th Gen NVIDIA NVLink Technology, and 4th Gen Tensor Cores with a new transformer engine
- 4x 3rd Gen NVIDIA NVSwitches for maximum GPU-GPU Bandwidth (7.2TB/sec of total BW) – Full all-to-all communication with 900GB/sec of bandwidth per GPU
- Supports GPUDirect® RDMA over PCI-E x16 5.0 to NVIDIA ConnectX®-7 NDR 400Gb InfiniBand adapters
- Up to 0 cores from 2x AMD EPYC 9005 Series CPUs
- Up to 6 TB system memory
- Additional slots for InfiniBand, 25/50/100G/200G/400G Ethernet, and high-speed storage
- NVIDIA-Certified Systems™ Configuration for the ultimate in performance and validation. NVIDIA AI Enterprise-ready.
- NVIDIA CUDA Toolkit installed and configured – ready to run GPU jobs!