Heterogeneous search in unstructured peer-to-peer networks

Abstract

Resource search or discovery is a fundamental issue in peer-to-peer (P2P) and grid studies. Search objects, or resources, can be cycles, storage spaces, files, services, addresses, and so on. In most search methods cited in related literature, peers run uniform forwarding logic; we call these homogeneous search (HMS) techniques. Although various efficient and scalable HMS techniques exist, previous studies show that the performance of any single method isn't perfect on all sides. Therefore, our research mixes those HMS techniques to increase performance; we call this heterogeneous search (HES). We explore the feasibility of HES evaluated in the context of timekeeping P2P networks from our prior studies. Our simulation results show HES to have significant advantages over other methods.