节点文献
非结构化P2P网络中一种新型的一跳查询与转发数据搜索新策略研究
A New Policy of 1-Hop Querying and Forwarding on Data Search in Unstructured P2P
【Author】 ZENG Bi-qing1+, CHEN Zhi-gang2 1(Department of Computer Engineering, NanHai Campus, South China Normal University, 528225, China) 2 (College of information Science and Engineering, Central South University, 410083, China)
【机构】 华南师范大学南海校区计算机工程系; 中南大学信息科学与工程学院;
【摘要】 为了提高非结构化P2P网络中数据查询搜索的效率,论文中提出了一种新型的一跳查询与转发数据搜索新策略(OHQFS),它是以非结构化P2P网络中的数据查询请求转发策略和非转发策略为基础,将它们整合、集成后得到的一种新策略。OHQFS策略中的查询源结点直接搜索其自身的所有邻居结点,并将查询请求转发给这些邻居结点,使得这些邻居结点再去搜索它们的相邻结点。该策略在数据查询搜索过程中无须维持一个很大的邻居结点信息集合,系统维护开销较小,通过本策略中固有的一步查询转发,使其邻居结点和邻居的邻居结点处于查询搜索范围内。网络仿真的实验结果表明,OHQFS策略相对于转发搜索策略而言,它提高了查询效率;而相对于非转发策略,它获得了较高的成功率。
【Abstract】 For the distributed P2P system, how to query the data efficiently and find the needed data from a mass of resources by the limited information is the key factor to the performance of P2P network system. Most searching policy in unstructured P2P are forwarding-based, including iterative deepening, local indices, k-walker random walk, modified random BFS, two-level k-walker random walk, directed BFS, intelligent search, routing indices based search, adaptive probabilistic search, and dominating set based search. In the forwarding-based policies, a querying task is forwarded on the overlay network from the querying source to the querying destinations where the desired data items are located. Once the query is forwarded, the querying source has no influence on the number of nodes that receive the query and in which order these nodes receive the query. The forwarding-based policy does not offer query flexibility and has low query efficiency. For non-forwarding policy, queries are not forwarded. Instead, the querying source has full control over query processing. Thus the querying source directly probes its neighbors for the data items it desires. In this paper, the forwarding-based and non-forwarding policies are combined to get their advantages while lowering their disadvantages. The research is to advocate the integration of both policies, and to evaluate the integrated policies. We propose a policy named 1-Hop Querying and Forwarding Search policy (OHQFS). The main idea of the OHQFS policy is the following: Given a query, the querying source directly probes its own extended neighbors for the desired files and forwards the query to a selected number of neighbors. These neighbors will probe their own extended neighbors on behalf of and under the guidance of the querying source and will not forward the query further. When the query termination condition is satisfied, the querying source terminates its own probing and the probing of its neighbors. It is a unification of direct query probing and guided 1-hop forwarding. The OHQFS policy is actually a 2-level tree rooted at the querying source. The querying source directly probes the nodes at level-1 of the tree. In the mean time, it also forwards the query to the internal nodes at level-1 and guides these nodes to probe the nodes at level-2 on its behalf. The querying source terminates the search when the query is satisfied or when all its neighbors and its neighbors’ neighbors are probed. The maximum searching scope for a query in this policy is the 2-hop neighborhood of the querying source. The performance of the OHQFS policy is evaluated experimentally against the forwarding-based policy and non-forwarding searching policy. The results of experiment shows: The higher success rate of the OHQFS policy is achieved at a query cost higher than the non-forwarding policy but lower than the forwarding policy. The experimental results demonstrate that the OHQFS policy combines the advantages of bot
- 【会议录名称】 第二届和谐人机环境联合学术会议(HHME2006)——第2届中国普适计算学术会议(PCC’06)论文集
- 【会议名称】第二届和谐人机环境联合学术会议(HHME2006)——第2届中国普适计算学术会议(PCC’06)
- 【会议时间】2006-10
- 【会议地点】中国浙江杭州
- 【分类号】TP393.02
- 【主办单位】清华大学计算机科学与技术系、浙江大学计算机科学与技术学院