A Spine-Leaf or "fat tree" architecture features multiple connections between interconnection switches (spine switches) and access switches (leaf switches) to support high-performance computer clustering. In addition to flattening and scaling out Layer 2 networks at the edge, it also creates a nonblocking, low-latency fabric.
[image]
Compared with a traditional three-tier architecture, Spine-Leaf uses fewer aggregation switches and redundant paths between the access and interconnection switches, thus reducing latency and energy requirements. This architecture is becoming the standard for greenfield data centers. All access switches are connected to every interconnection switch, providing a redundant path. Traffic runs east-west, reducing latency between access switches. This is a very scalable fabric.
In the example below, the interconnection switch is connected directly to the access switch with 12- or 24-fiber MTP trunks and either a conversion module (24-fiber) or MTP adapter panel (12-fiber).
[image]
Look to Leviton for an ideal cabling infrastructure for Spine-Leaf designs. Our HDX Unity conversion cassettes easily snap in and out of UHDX panels or enclosures, while the Opt-X Unity system of trunks, harnesses, and cords make it easy to migrate from 10 to 40, 100, 200 and 400 G/bs networks to speed tech refreshes, reduce labor and minimize network downtime.
