Persistent Last-mile Congestion

Abstract

Last-mile is the centerpiece of broadband connectivity, as poor last-mile performance generally translates to poor quality of experience. In this work we investigate last-mile latency using traceroute data from RIPE Atlas probes located in 646 ASes and focus on recurrent performance degradation. We find that in normal times probes in only 10% ASes experience persistent last-mile congestion but we recorded 55% more congested ASes during the COVID-19 outbreak. Persistent last-mile congestion is not uncommon, it is usually seen in large eyeball networks and may span years. With the help of CDN access log data, we dissect results for major ISPs in Japan, the most severely affected country in our study, and ascertain bottlenecks in the shared legacy infrastructure.

KL-Dection: An Approach to Detect Network Outages Based on Key Links

“…RTT-based outage detection approaches, such as [1,4,6,15], have been proposed by utilizing different statistical models to characterize the RTT of all links to detect the network outages. Fontugne et al [1] first obtained the differential RTT of all links.…”

“…RTT-based outage detection approaches, such as [1,4,6,15], have been proposed by utilizing different statistical models to characterize the RTT of all links to detect the network outages. Fontugne et al [1] first obtained the differential RTT of all links. Then, they leveraged normal distribution to model the measurements and detected the network outages by applying the Wilson score.…”

“…The unprecedented growth of the Internet has resulted in an explosive increase in network security issues, such as network outages. Network outages inevitably degrade network connectivity and influence network performance [1][2][3][4]. For example, the network outage caused by censorship in 2011 blocked the Internet access of Libya [5].…”

“…How-ever, they lead to high resource occupancy and cost in practice. The reasons lie in the following: (1) monitoring the network performance of all links needs to perform a bulk of measurement tasks. These tasks will inject extra traffic into the network, which may occupy the link bandwidth, reduce the network transmission speed, and increase the network burden [8,9].…”

WebRTC-based measurement tool for peer-to-peer applications and preliminary findings with real users

“…Home Area Latency Measurement 2 aimed to provide a service that estimates the latency of a home network by calculating the difference between the RTT from a home node to a measurement server and the measurement server to the exit point of the home network. Fontugne et al revealed congestion and communication bottlenecks in the last mile by analyzing the traceroute data from RIPE Atlas in [8].…”