If you are designing a configuration for a large SAN, you must estimate the queue depth for each node to avoid application failures.
The queue depth is the number of I/O operations that can run in parallel on a device.
If asystem node reaches the maximum number of queued commands, many operating systems cannot recover if the situation persists for more than 15 seconds. This situation can result in one or more servers that present errors to applications and application failures on the servers.
A large SAN is one in which the total number of host mappings is at least 1000. For example, 50 servers with each server addressing 20 volumes.