NCP-AIN Exam Dumps : NVIDIA-Certified Professional AI Networking

EVPN Multi-homingenablesactive-active redundancy without inter-switch linksby usingoverlay routingover VXLAN and distributed control plane using BGP EVPN.

From the officialNVIDIA Cumulus Linux EVPN Multihoming Documentation:

"EVPN multihoming allows multiple Top-of-Rack (ToR) switches to connect to the same server while maintaining full layer-2 redundancy without the need for inter-switch links or traditional MLAG configuration."

Key benefits:

Simplified topology (no ISL/peer-link needed)

BGP-based control plane

Fast convergence

Active-active links per host NIC

Incorrect Options:

MLAGrequires ISL between switches and peer-link configuration.

VSS(Virtual Switching System) is a Cisco term, not supported in NVIDIA networking.

MLNX_OFED (Mellanox OpenFabrics Enterprise Distribution) is an NVIDIA-tested and packaged version of the OpenFabrics Enterprise Distribution (OFED) for Linux. It provides the necessary drivers and tools to support InfiniBand and Ethernet interconnects using the same RDMA (Remote Direct Memory Access) and kernel bypass APIs. MLNX_OFED enables high-performance networking capabilities essential for AI clusters, including support for up to 400Gb/s InfiniBand and RoCE (RDMA over Converged Ethernet).​

Reference Extracts from NVIDIA Documentation:

"MLNX_OFED is an NVIDIA tested and packaged version of OFED that supports two interconnect types using the same RDMA (remote DMA) and kernel bypass APIs called OFED verbs – InfiniBand and Ethernet."​

"Up to 400Gb/s InfiniBand and RoCE (based on the RDMA over Converged Ethernet standard) over 10/25/40/50/100/200/400GbE are supported."​

For optimal performance in multi-node, multi-GPU AI workloads:

NVLinkprovides high-speed, low-latency communication between GPUs within the same node.

InfiniBandoffers efficient, scalable communication between nodes in a data center.Combining these technologies ensures both intra-node and inter-node communication needs are effectively met.​