As you plan your MetroCluster configuration, you should understand the hardware components and how they interconnect.
A fabric MetroCluster configuration consists of one or two DR groups, depending on the number of nodes in the MetroCluster configuration. Each DR group consists of four nodes.
The following illustration shows the organization of nodes in an eight-node MetroCluster configuration:
The following illustration shows the organization of nodes in a four-node MetroCluster configuration:
A MetroCluster configuration includes the following key hardware elements:
The storage controllers are not connected directly to the storage but connect to two redundant FC switch fabrics.
The FC-to-SAS bridges connect the SAS storage stacks to the FC switches, providing bridging between the two protocols.
The FC switches provide the long-haul backbone ISL between the two sites. The FC switches provide the two storage fabrics that allow data mirroring to the remote storage pools.
The cluster peering network provides connectivity for mirroring of the cluster configuration, which includes storage virtual machine (SVM) configuration. The configuration of all of the SVMs on one cluster is mirrored to the partner cluster.
An eight-node configuration consists of two clusters, one at each geographically separated site. cluster_A is located at the first MetroCluster site. cluster_B is located at the second MetroCluster site. Each site has one SAS storage stack. Additional storage stacks are supported, but only one is shown at each site. The HA pairs are configured as switchless clusters, without cluster interconnect switches. A switched configuration is supported, but is not shown.
An eight-node configuration includes the following connections:
If the controllers support a single-chassis HA pair, the HA interconnect is internal, occurring through the backplane, meaning that an external interconnect is not required.
SVM configuration is replicated over the cluster peering network.
The following illustration shows a simplified view of a four-node fabric MetroCluster configuration. For some connections, a single line represents multiple, redundant connections between the components. Data and management network connections are not shown.
The following illustration shows a more detailed view of the connectivity in a single MetroCluster cluster (both clusters have the same configuration):
The following illustration shows a simplified view of a two-node fabric MetroCluster configuration. For some connections, a single line represents multiple, redundant connections between the components. Data and management network connections are not shown.
A two-node configuration consists of two clusters, one at each geographically separated site. cluster_A is located at the first MetroCluster site. cluster_B is located at the second MetroCluster site. Each site has one SAS storage stack. Additional storage stacks are supported, but only one is shown at each site.
The following illustration shows a more detailed view of the connectivity in a single MetroCluster cluster (both clusters have the same configuration):
A two-node configuration includes the following connections:
SVM configuration is replicated over the cluster peering network.