P2P (peer-to-peer) is the key technology of reconstructing the future distributed architecture and has a good application perspective. As the issues in P2P systems mostly come down to data placement and retrieval, P2P data management has recently become an active topic in database community. In this paper, the advantages of P2P systems are first described. Then the goals of P2P data management researches are presented. Thirdly, research of P2P data management is described from three facets, i.e. P2P information retrieval, P2P database-style queries and P2P continuous queries. Particularly, the index construction methods, semantic coordination, query semantics, query processing strategies, types of queries supported, and query optimization of P2P database-style queries are discussed in detail. Finally, the issues to be further studied are proposed.