N10-009 Question 106
Single answerMesh, Hybrid, Star/hub and spoke, Spine and leaf, Point to pointA company is building a new virtualized data center for an internal private cloud. The environment will host hundreds of east-west traffic flows between application servers, web servers, and storage services. The network team wants a design that provides predictable low-latency paths between racks, scales horizontally as new racks are added, and avoids the bottlenecks common in a traditional three-tier architecture. Which topology is the BEST fit for this requirement?
- A
Star (hub-and-spoke) topology
- B
Spine-and-leaf topology
- C
Point-to-point topology
- D
Full mesh topology
Show answer and explanation
Correct answer: B
Explanation
The best answer is spine-and-leaf topology because it is purpose-built for modern data center traffic patterns, especially east-west communication between servers and services. In a spine-and-leaf fabric, each leaf switch connects upstream to every spine switch, which creates deterministic forwarding behavior and usually a consistent number of hops between hosts in different racks. This supports low latency, high bandwidth, and horizontal scaling as additional racks are deployed. By contrast, star/hub-and-spoke designs are more common for centralized branch connectivity, point-to-point links connect only two endpoints, and full mesh topologies do not scale efficiently in large switching environments. This aligns with common enterprise data center design best practices described by major networking vendors and data center architecture guidance, which recommend spine-and-leaf for scalable, high-performance server fabrics.
- A. Incorrect.
Incorrect. A star, or hub-and-spoke, design centralizes connectivity through an aggregation point. While it is simple to manage and common in branch/WAN environments, it can create bottlenecks when large amounts of east-west traffic must traverse central devices. In a modern data center with frequent server-to-server communication, this is usually less efficient than a spine-and-leaf design.
- B. Correct.
Correct. Spine-and-leaf is specifically designed for data center environments with heavy east-west traffic. Each leaf switch connects to every spine switch, creating predictable paths with consistent hop counts between endpoints. This architecture scales well by adding additional leaf or spine switches and is widely used to reduce oversubscription and improve throughput compared with traditional hierarchical designs.
- C. Incorrect.
Incorrect. Point-to-point topology refers to a dedicated link between two nodes. It is useful for simple direct connections such as leased lines or inter-device links, but it does not describe a scalable data center switching fabric for many racks and many-to-many communication patterns.
- D. Incorrect.
Incorrect. A full mesh provides excellent redundancy because every node connects to every other node, but it becomes impractical and expensive at data center scale due to the large number of links and interfaces required. Candidates may choose this because mesh implies high availability, but spine-and-leaf achieves high redundancy and scalability in a far more practical way for modern data centers.