It is a common term in IP network monitoring, and differs from round-trip time (RTT) in that only path in the one direction from source to destination is measured.
The ping utility measures the RTT, that is, the time to go and come back to a host.
Half the RTT is often used as an approximation of OWD but this assumes that the forward and back paths are the same in terms of congestion, number of hops, or quality of service (QoS).
To avoid such problems, the OWD may be measured directly.
OWDs may be measured between two points A and B of an IP network through the use of synchronized clocks; A records a timestamp on the packet and sends it to B, which notes the receiving time and calculates the OWD as their difference.
However, this method suffers several limitations, such as requiring intensive cooperation between both parties, and the accuracy of the measured delay is subject to the synchronization precision.
[1] Transmission between two network nodes may be asymmetric, and the forward and reverse delays are not equal.
The accuracy of such an estimate depends on the nature of delay distribution in both directions.
As delays in both directions become more symmetric, the accuracy increases.
The probability mass function (PMF) of absolute error, E, between the smaller of the forward and reverse OWDs and their average (i.e., RTT/2) can be expressed as a function of the network delay distribution as follows:[1]