Articles by tag "HPC networking"

1. Cost-Effective Deployment of NVIDIA Mellanox Switches for High-Bandwidth AI Clusters
   Selene Gong

   36
   High-bandwidth AI clusters rely on low-latency, high-throughput networking to support demanding workloads such as machine learning training, inference, and large-scale simulations. NVIDIA Mellanox switches deliver high port density, scalable throughput, and robust reliability. This guide ...
2. Troubleshooting InfiniBand Subnet Manager Failover High-Availability OpenSM Architecture for MQM8790 Unmanaged Clusters
   Selene Gong

   18
   Quick Take In deployments based on unmanaged MQM8790 switches, the entire InfiniBand fabric control plane relies on an external Subnet Manager (SM). Implementing a high-availability active/standby OpenSM architecture with distinct sm_priority values prevents full cluster outages during ...
3. NVIDIA Quantum-2 MQM9790-NS2F Specifications & Procurement Guide: Unlocking 400G NDR for AI Clusters
   Selene Gong

   18
   Quick Take The NVIDIA Quantum-2 MQM9790-NS2F is a pioneering 64-port NDR 400Gb/s InfiniBand switch that delivers 51.2 Tbps of non-blocking capacity in a 1U chassis. By leveraging an innovative 32 twin-port OSFP cage design, ultra-low latency forwarding, and hardware-driven enhanced SHARP ...