作者: David R. Choffnes , Mario A. Sanchez , Fabian E. Bustamante
DOI: 10.1109/INFCOM.2010.5462225
关键词: Computer network 、 Hybrid positioning system 、 Computer science 、 Empirical research 、 Robustness (computer science) 、 The Internet 、 Network simulation
摘要: Network positioning systems provide an important service to large- scale P2P systems, potentially enabling clients achieve higher performance, reduce cross-ISP traffic and improve the robustness of system failures. Because traces representative this environment are generally unavailable, there is no platform suited for experimentation at appropriate scale, network have been commonly imple- mented evaluated in simulation on research testbeds. The performance remains open question large deployments edges network. This paper evaluates how four key classes po- sitioning fare when deployed measured where they used. Using 2 billion measurements gathered from more than 43,000 IP addresses probing over 8 million other IPs worldwide, we show that exhibits noticeably worse previously reported studies conducted To explain result, identify several properties call into fundamental assumptions driving research.