Home > Storage > PowerStore > Virtualization and Cloud > Dell PowerStore: Microsoft Hyper-V Best Practices > Queue depth best practices for Hyper-V
Queue depth refers to the number of disk transactions that can be in flight from an initiator port (on a host server) to a target port (on the storage array). Host server FC and iSCSI adapters have queue depth settings that can be modified.
A target port on PowerStore storage supports multiple host initiator ports sending it data concurrently. Initiator queue depth is used to limit the number of transactions an initiator can send to a target. Flooding occurs when a target port becomes saturated, and transactions are queued. Flooding causes higher latency and degraded performance for the affected workloads.
With PowerStore SAN configurations, configure up to four front-end data (target) ports per PowerStore node per fabric. Configuring more than four ports per PowerStore node per fabric might cause performance degradation. Configure two to four ports for an optimal configuration. Using multiple target ports allows I/O to be spread out, reducing the risk of port saturation.
On a Windows Server host, queue depth is a function of the Microsoft storport.sys driver and the vendor-specific miniport driver for the FC or iSCSI adapter. Default queue-depth settings provide a good starting point and are adequate for most workloads.
Note: We do not recommend modifying queue-depth settings unless there is a specific reason to do so. Test any queue-depth changes before applying them in a production environment.
Consider the following example:
However, consider the possible negative impact if many hosts are mapped to this storage array.
See the documentation for your host adapter for information about adjusting queue depth settings.
For example, see the Marvell QLogic Fibre Channel Adapters Users Guide at Marvell.com.