A user has submitted a trouble ticket complaining that the email server VM has become very sluggish. After looking at the logs, the vSphere administrator suspects the iSCSI network is experiencing con

GAVG (Guest Average) represents the total I/O latency observed by the guest OS and is calculated as KAVG (kernel latency) plus DAVG (device/storage latency). GAVG/wr tracks this end-to-end write latency specifically, so elevated values directly confirm that the guest is waiting longer for storage writes - consistent with iSCSI network congestion adding device-side delay. This metric is the most meaningful indicator of actual guest OS performance impact.

%ACTV shows the percentage of time commands are active on a storage device, which reflects utilization rather than the latency experienced by the guest OS.

PATH displays multipath topology and state information for storage adapters, not a performance or latency measurement.

PKTRXMUL/s is a network statistic counting multicast packets received per second and has no direct relationship to storage I/O latency or guest OS performance.