| D. Kossman. The state of the art in distributed queyr processing. ACM Computing Survey, September 2000. |
.... simulations, and find that RIs can improve performance by one or two orders of magnitude over a flooding based system, and by 50 100 versus a random forwarding system (Section 7) 2 Related Work The problem of indexing a P2P network is related to the problem of indexing a distributed database [13]. However, algorithms for indexing distributed databases make two fundamental assumptions that are not applicable to P2P systems: that nodes are stable and connected most of the time, and that the number of nodes is small. There are several working P2P systems currently available, each with its ....
D. Kossman. The state of the art in distributed queyr processing. ACM Computing Survey, September 2000.
....SONs, an implementation of SONs that further improves query performance at the expense of a marginal reduction of the maximum achievable recall level. 2. RELATED WORK The idea of placing data in nodes close to where relevant queries originate was used in early distributed database systems [5]. However, the algorithms used for distributed databases are based on two fundamental assumptions that are not applicable to P2P systems: that there are a small number of stables nodes, and that the designer has total control over the data. There are a number of P2P research systems (CAN [10] ....
D. Kossman. The state of the art in distributed queyr processing. ACM Computing Survey, September 2000.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC