Explanation/Reference:
Section: Enterprise Network Design Explanation
Explanation:
Of the choices available, the problem with the network design in the diagram is that not every distribution layer device has multiple paths to the core layer. Hierarchical design separates functionality into distinct layers, providing for more efficient performance and greater scalability. The hierarchical design model divides the network into three distinct layers: the core layer, the distribution layer, and the access layer The core layer provides fast transport services and redundant connectivity to the distribution layer.
Because the core layer acts as the network's backbone, it is essential that every distribution layer device have multiple paths to the core layer. Multiple paths between the core and distribution layer devices ensure that network connectivity is maintained if a link or device fails in either layer. The network design shown in the diagram above can be improved with the addition of a second link from the central distribution layer switch to one of the alternate core layer switches, as shown by the green line in the diagram below:

Although a fully meshed topology can be implemented in the core layer, a fully meshed topology is not required if multiple paths exist between core layer and distribution layer devices. In addition, you should not implement a fully meshed topology in the access layer because the access layer should remain highly scalable. Because a fully meshed topology can add unnecessary cost and complexity to the design and operation of the network, a partially meshed topology is often implemented in the core layer. In this scenario, creating a fully meshed topology in the core layer would not improve the network design, because the central distribution layer switch would still have only a single connection into the core layer.
The network diagram in this scenario indicates that Layer 3 switching is implemented in the core and distribution layers. Although Layer 3 switching is the preferred switching mechanism in the core layer, Layer 3 switching is not required. Layer 3 switching is preferred in the core layer because it can provide faster convergence times than Layer 2 switching can provide after a link failure or a device failure. In addition, Layer 3 switching in the core layer facilitates the implementation of load balancing and path optimization.
There is not enough information in the diagram to determine whether the core layer contains a sufficient number of devices. The core layer should contain enough devices to provide wire-speed transport services to the distribution layer devices. In addition, the core layer should contain enough devices to support redundant paths to each distribution layer device.
Reference:
Cisco: Campus Network for High Availability Design Guide: Core Layer