IP Fabric Architectures (Spine/Leaf) JN0-281 Practice Questions (Page 3)

Questions 11–15

1. 11foundation · easy

   What is a single point of failure in a network?

   • AA component that is not redundant but can be easily replaced.
   • BA component that is shared by multiple tenants.
   • CA component whose failure would cause the entire network to become unavailable.
   • DA component that is used only for management purposes.
   Reveal Solution Discussion Select an answer first
2. 12foundation · easy

   In a spine-leaf fabric, what does ECMP use to distribute traffic across multiple spines?

   • ARound-robin scheduling based on packet arrival time.
   • BThe lowest latency path at the moment of forwarding.
   • CThe path with the least amount of traffic.
   • DA hash of packet headers to select a path.
   Reveal Solution Discussion Select an answer first
4. 13expert · hard

   A large enterprise is designing a spine-leaf fabric for a new data center. The workload is primarily east-west traffic between application tiers. The network team has a requirement to maximize throughput between any two servers. They are considering two designs: Design A uses 4 spines with 40GbE uplinks, and Design B uses 8 spines with 25GbE uplinks. Both designs have the same number of leaf switches. Which design will provide higher aggregate bandwidth between any two leaves?

   • ADesign A, because 4 spines with 40GbE uplinks provide a total of 160GbE of uplink bandwidth per leaf.
   • BBoth designs provide the same aggregate bandwidth because the total number of spine ports is the same.
   • CThe aggregate bandwidth is the same because the number of leaves is the same in both designs.
   • DDesign B, because 8 spines with 25GbE uplinks provide a total of 200GbE of uplink bandwidth per leaf.
   Reveal Solution Discussion Select an answer first
5. 14application · medium

   A company is comparing a traditional three-tier network (core/aggregation/access) with a spine-leaf fabric for a new data center. The application workload is primarily east-west traffic between servers. The network architect needs to justify the spine-leaf design based on latency and scalability. Which comparison is accurate?

   • ASpine-leaf requires more cabling than a three-tier design, but this is offset by the ability to use Spanning Tree Protocol for loop prevention.
   • BSpine-leaf provides a maximum of two hops between any two servers, while a three-tier design can require up to six hops for server-to-server traffic.
   • CA three-tier design is simpler to operate because it uses a single routing protocol, while spine-leaf requires multiple protocols.
   • DThree-tier networks are inherently more scalable because they use larger, more expensive chassis switches at the core.
   Reveal Solution Discussion Select an answer first
6. 15application · medium

   A data center is running a spine-leaf fabric with 4 spines and 10 leaves. The network team needs to increase the total bandwidth available between any two leaves. They want to avoid any disruption to the existing traffic. Which action will achieve this goal?

   • AAdd a new leaf switch and connect it to all existing spine switches.
   • BIncrease the number of uplinks from each leaf to a single spine switch.
   • CReplace all existing spine switches with models that have higher port speeds.
   • DAdd a new spine switch and connect it to all existing leaf switches.
   Reveal Solution Discussion Select an answer first