Summary

There is considerable industry interest in conducting tests that capture latency of the user's path while in‑use. The traffic load level associated with "while in-use" is widely proportional to the additional latency measured with respect to the minimum latency, and latency increases rapidly near the maximum load level. A user using application traffic will experience latency within measurable bounds: the lower limit of the minimum latency and the upper limit of latency at maximum capacity.

Although many measurement applications available at the time of writing assess different metrics of latency, they mostly employ the TCP protocol. Therefore, user datagram protocol (UDP) based latency under load/responsiveness is a clear gap. This gap is significant, because:

1.          User applications/traffic with strict response time requirements will use UDP.

2.          The percentage of UDP traffic is significant and continues to grow in proportion to TCP traffic.

Recommendation ITU-T Y.1567 specifies metrics of latency under simultaneous traffic load and defines methods of measurement to increase the specificity and repeatability of metric assessment.