This topic describes the rules and requirements for configuring a SAN Volume Controller. It also provides a list of defined terms that are referenced in the configuration rules. Before you read the rules, read these definitions, which can help you to understand the rules.
With reference to all pairs of N-ports or end-nodes that are in a fabric, an ISL hop is the number of links that are crossed on the shortest route between the node pair whose nodes are farthest apart from each other. The distance is measured only in terms of the ISL links that are in the fabric.
This definition assumes a symmetrical network and a specific workload that is applied equally from all initiators and sent equally to all targets. A symmetrical network means that all initiators are connected at the same level and all the controllers are connected at the same level.
The SAN Volume Controller makes this calculation difficult, because it puts its back-end traffic onto the same network, and this back-end traffic varies by workload. Therefore, the oversubscription that a 100% read hit gives is different from the oversubscription that 100% write-miss gives.
If you have an oversubscription of 1 or less, the network is nonblocking.
Because the SAN Volume Controller supports Remote Copy, significant distances might exist between the components of the local cluster and those of the remote cluster.
Because the SAN Volume Controller supports remote copy, significant distances might exist between the components of the local cluster and those of the remote cluster.
Some controllers recommend that the number of hosts that use each port be limited to prevent excessive queuing at that port. If the port fails, or the path to that port fails, the host might failover to another port, and the fan-in requirements might be exceeded in this degraded mode.
Usually this type of configuration does not create an error log.
Typically, a repair action is required to restore the degraded configuration to a valid configuration.