A two-tier network design used in data centers that makes latency easy to keep uniform.
A data-center network design consisting of two tiers, spine switches and leaf switches, where any leaf can reach any other leaf in the same number of hops. With fewer tiers than the traditional three-tier core/distribution/access design, it handles East-West traffic with low, uniform latency, which is why it's widely adopted in today's virtualized, microservices-heavy data centers.
© 2026 ITBGM