作者: Shen Tat Goh , Panos Kalnis , Spiridon Bakiras , Kian-Lee Tan
DOI: 10.1007/11408079_19
关键词:
摘要: The fundamental drawback of unstructured peer-to-peer (P2P) networks is the flooding-based query processing protocol that seriously limits their scalability. As a result, significant amount research work has focused on designing efficient search protocols reduce overall communication cost. What lacking, however, availability real data, regarding exact content users' libraries and queries these users ask. Using trace-driven simulations will clearly generate more meaningful results further illustrate efficiency generic under real-life scenario. Motivated by this fact, we developed Gnutella-style probe collected detailed data over period two months. They involve around 4,500 contain files shared each user, together with any available metadata (e.g., artist for songs) information about nodes connection speed). We also initiated users. After filtering, were organized in XML format are to researchers. Here, analyze dataset present its statistical characteristics. Additionally, as case study, employ it evaluate recently proposed P2P searching techniques.