End-to-End Latency Measurements with Black Box Components

Edge computing has gained substantial attention in both academia and industry due to its ability to facilitate cloud services provision in close proximity to the user and data source, thereby effectively reducing latency. This is particularly important for latency-sensitive or tactile Internet services that cannot tolerate delays caused by transferring workloads to distant cloud computing data centers. User equipment generating workloads for offloading to the edge encompasses a wide range of device types, from small sensors to specialized integrated devices like augmented reality wearables, and even powerful mobile computing devices like smartphones, tablets, or laptops. Since low latency is the foremost factor for utilizing edge computing services, it is imperative to identify and resolve any latency problems that might occur between the user equipment and the edge server. Nevertheless, accurately measuring latency becomes a challenging task when the edge service setting comprises closed black box components or entire systems. For instance, closed user equipment, such as augmented reality wearables, typically do not allow intrusive measurement techniques, and synchronization with other components of the setting is often not possible. To address these challenges, this article proposes a pragmatic end-to-end latency measurement approach specifically designed for edge computing settings with black box components. This approach leverages performance data provided by the Real-Time Control Protocol (RTCP) to obtain accurate latency measurements. The results obtained using this approach are validated through reference scenarios with only white box components, and exemplary latency measurements within an edge computing setting deploying a private 5G network infrastructure are presented.