You can configure a Cisco UCS instance to use a single fabric interconnect in a standalone configuration or to use a redundant pair of fabric interconnects in a cluster configuration.
A cluster configuration provides high availability. If one fabric interconnect becomes unavailable, the other takes over. Only one management port (Mgmt0) connection is required to support a cluster configuration; however, both Mgmt0 ports should be connected to provide link-level redundancy.
In addition, a cluster configuration actively enhances failover recovery time for redundant virtual interface (VIF) connections. When an adapter has an active VIF connection to one fabric interconnect and a standby VIF connection to the second, the learned MAC addresses of the active VIF are replicated but not installed on the second fabric interconnect. If the active VIF fails, the second fabric interconnect installs the replicated MAC addresses and broadcasts them to the network through gratuitous ARP messages, shortening the switchover time.
The cluster configuration provides redundancy only for the management plane. Data redundancy is dependent on the user configuration and may require a third-party tool to support data redundancy.
To use the cluster configuration, the two fabric interconnects must be directly connected together using Ethernet cables between the L1 (L1-to-L1) and L2 (L2-to-L2) high availability ports, with no other fabric interconnects in between. This allows the two fabric interconnects to continuously monitor the status of each other and quickly know when one has failed.
Both fabric interconnects in a cluster configuration must go through the initial setup process. The first fabric interconnect to be set up must be enabled for a cluster configuration. Then, when the second fabric interconnect is set up, it detects the first fabric interconnect as a peer fabric interconnect in the cluster.
For more information, refer to the Cisco UCS 6100 Series Fabric Interconnect Hardware Installation Guide.