Internet engineering, science, and public policy communities have significant interest in understanding the extent and scope of, as well as potential consumer harm induced by, persistent interdomain congestion on the Internet. In recent work (Dhamdhere et al., 2018), we developed and implemented a lightweight active measurement method and system to measure evidence of congestion on thousands of interconnection links between broadband access ISPs and major interconnecting parties, including directly connected content providers. This method provides empirical grounding for discussions of interconnection congestion, without requiring direct access to interconnection links.

We first review the previous work to provide context. We then present new techniques for visualizing the data in ways we believe are conducive to policy analysis, e.g, of infrastructure resilience, performance metrics, and potential harm to consumers of persistently under-provisioned interconnection links. We present data in ways that allow us to compare different access providers, and show how congestion varies over time. We focus on seven large U.S. broadband access networks, but there is nothing U.S.-specific about the methods we use. Finally, we describe policy-relevant limitations and implications of the work.