Committee: Reza Rejaie (chair), Kevin Butler, Jun Li
Directed Research Project(Aug 2013)
Keywords: content providers; PoP level topology; Active Measurement
Large content providers (CPs) are responsible for a large fraction of injected traffic to the Internet. They maintain multiple data centers and they connect to different ASes to relay their contents and service traffic to the rest of the Internet. In this paper we propose a novel methodology to measure and characterize large Internet Content Providers. Basically our contribution is two-fold:
- We design a targeted CP oriented measurement methodology. This methodology includes Application Level Probing method, using name aliasing, subnet discovery and optimized tracerout probing.
- We show how to clean and aggregate the collected data to effectively characterize the target CP, finding PoP locations and explaining routing policies.
As a case study we apply this methodology on Google Maps to show the feasibility of our method. Considering the constraint view of this methodology, we discuss how it is capable of discovering more peering relations than other general large scale measurements and how it unravels the routing mechanisms of Google Maps service.