Internet Telescope

Internet is a web of interconnected backbone networks of high speed. Thousands of small and medium size Autonomous Systems (ASes) interconnect individuals, businesses, universities, and agencies over the Internet backbone. Internet is the largest man-made complex network which has far exceeded initial design goals. While the building blocks of the Internet, its protocols and individual components, have been subject to intensive studies, the immense global entity has not been precisely characterized. Moreover, the Internet's global properties cannot be inferred from local ones as it is composed of networks engineered with large technical diversity and range from small local campuses to large transcontinental backbone providers. There is a general lack of understanding of Internet topology.

Understanding the topological and the functional characteristics of the Internet is an important research issue as the Internet grows with no central authority. This understanding is not simply an intellectual curiosity, but also a necessity to better design, implement, protect and operate the underlying network technologies, protocols, and services. Complexity is a major constraint for the design and management of computer networks and protocols.

In this project, we develop an Internet Telescope, a system that provides insight into Internet topology by taking snapshots of the underlying networks. Internet Telescope will collect topology information from the Internet using PlanetLab. The system utilizes efficient algorithms to process large scale data-sets collected from various vantage points and provide topology graphs. The Internet Telescope will help to (1) fine-tune existing services such as content distribution, (2) develop more efficient protocols, (3) monitor connectivity and identify bottlenecks, (4) and guide the development of the next generation Internet.

Below are two related slides.

• Internet Topology Discovery
• Internet Measurements

People:

• Mehmet H. Gunes (University of Nevada, Reno)
• Kamil Sarac (University of Texas at Dallas)
• Mehmet Engin Tozal (University of Texas at Dallas)
• Talha Oz (University of Nevada, Reno)
• David Shelley (University of Nevada, Reno)

Publications:

• Mehmet H. Gunes and Kamil Sarac, "Resolving IP aliases in Building Traceroute-Based Internet Maps", IEEE/ACM Transactions on Networking (to appear).
  
  Abstract: Alias resolution, the task of identifying IP addresses belonging to the same router, is an important step in building traceroute-based Internet topology maps. Inaccuracies in alias resolution affect the representativeness of constructed topology maps. This in turn affects the conclusions derived from studies that use these maps. This paper presents two complementary studies on alias resolution. First, we present an experimental study to demonstrate the impact of alias resolution on topology measurement studies. Then, we introduce an alias resolution approach called Analytic and Probe-based Alias Resolver (APAR). APAR consists of an analytical component and a probe-based component. Given a set of path traces, the analytical component utilizes the common IP address assignment scheme to infer IP aliases. The probe-based component introduces a minimal probing overhead to improve the accuracy of APAR. Compared to the existing state-of-the-art tool ally, APAR uses an orthogonal approach to resolve a large number of IP aliases that ally fails to identify. Our extensive verification study on sample data sets shows that our approach is effective in resolving many aliases with good accuracy. Our evaluations also indicate that the two approaches (ally and APAR) should be used together to maximize the success of the alias resolution process.
  
  
• Mehmet H. Gunes and Kamil Sarac, "Analyzing Router Responsiveness to Measurement Probes", Passive and Active Measurement Conference (PAM 2009), Seoul, South Korea, April 1-3 2009.
  
  Abstract:Active probing has increasingly been used to collect information about the topological and functional characteristics of the Internet. Given the need for active probing and the lack of a widely accepted mechanism to minimize the overhead of such probes, the tra±c and processing overhead introduced on the routers are believed to become an important issue for network operators. In this paper, we conduct an experimental study to understand the responsiveness of routers to active probing both from a historical perspective and current practices. One main finding is that network operators are increasingly configuring their devices not to respond to active direct probes. In addition, ICMP based probes seem to elicit most responses and UDP based probes elicit the least.
  
  

Last Updated on Nov 9, 2009.