Networking / Communications

Supporting Server Redundancy with NVIDIA EVPN Multihoming

AI-Generated Summary

  • EVPN multihoming (EVPN-MH) is the latest addition to the NVIDIA EVPN story and provides support for all-active server redundancy.
  • A three-part video series walks through the design elements of EVPN-MH, starting with a feature overview and comparison with EVPN-MLAG.
  • The second video covers unicast packet flows including new data plane constructs such as MAC-ECMP and layer-2 nexthop-groups introduced for EVPN-MH.
  • The third video explains how PIM-SM optimizes flooded traffic, covering DF election and split horizon filtering.

Next Steps

Powered by NVIDIA Nemotron. AI-generated content may summarize information incompletely. Verify important information. Learn more

This post was originally published on the Cumulus Networks site.

EVPN multihoming (EVPN-MH) is the latest addition to the NVIDIA EVPN story. In this three-part video series, I walk you through the various design elements of EVPN-MH. 

Part 1: EVPN multihoming vs. EVPN with MLAG

EVPN-MH provides support for all-active server redundancy. In this video, I offer a feature overview and a comparison with EVPN-MLAG.

Video 1. EVPN MH overview.

Part 2: EVPN-MH: Packet flows

In this video, I dive into the various unicast packet flows in a network with EVPN-MH. This includes new data plane constructs such as MAC-ECMP and layer-2 nexthop-groups that have been introduced for the express purpose of EVPN-MH. 

Video 2. EVPN MH packet flow.

Part 3: EVPN-MH: Bum traffic

PIM-SM is used for optimizing flooded traffic in network with EVPN-MH. In this video, I walk through the implementation aspects of flooded traffic, including DF election and split horizon filtering.

Video 3. EVPN MH flooded traffic.

If you’d like to see more videos like these or more information on EVPN-MH, let me know in the comments.

Discuss (0)

Tags