Which characteristic describes spine-and-leaf architecture?

Spine-and-leaf architecture is designed for data centers to provide high-bandwidth, low-latency, and non-blocking network access. A key characteristic is that every device (server or endpoint) connected to a leaf switch is always the same number of hops away from any other device in the network. This consistent hop count ensures predictable latency and simplifies network design and troubleshooting. The other options are incorrect: "Each link between leaf switches allows for higher bandwidth" is incorrect because leaf switches do not directly connect to each other; they connect only to spine switches. "It provides greater predictability on STP blocked ports" is incorrect because spine-and-leaf architectures typically use technologies like Equal-Cost Multi-Path (ECMP) routing instead of Spanning Tree Protocol (STP) to avoid blocked ports and utilize all available links. "It provides variable latency" is incorrect because the consistent hop count is specifically designed to provide predictable and low latency, not variable latency.