作者: Ingo Brunkhorst , Hadhami Dhraief , Alfons Kemper , Wolfgang Nejdl , Christian Wiesner
DOI: 10.1007/978-3-540-24629-9_14
关键词:
摘要: Databases have employed a schema-based approach to store and retrieve structured data for decades. For peer-to-peer (P2P) networks, similar approaches are just beginning emerge, also motivated by the fact, that sending (atomic) queries appropriate peers clearly fails which need from more than one peer be executed. While quite few database techniques can re-used in this new context, P2P management infrastructure poses additional challenges solved before networks become as common databases. Because of dynamic nature we neither assume global knowledge about distribution, nor static topologies query plans suitable these networks. Unlike traditional distributed systems, cannot complete schema instance but rather work with directs processing tasks node or neighboring nodes.