Consolidated Review of From Paris to Tokyo: On the Suitability of Ping to Measure Latency

1. Strengths This paper makes the point that load balancers break ping just as they do traceroute. In fact, the authors discover that, due to link aggregation, even measurements along the same IP-level path may have very different delays. The paper proposes to apply the Paris-traceroute idea of controlling path identifier to ping, resulting in Tokyo-ping. It then examines this approach to show the problems that occur without it. A good problem to work on: ping is widely used, and the paper helps us interpret its results and improve the tool. Very suitable for a short paper. Paper goes on to address many of the questions that I had while reading it. Interesting conclusion: "high variability ...is likely a measurement artifact of the [ping] tool itself." Well-written and interesting to read.