Research Of Continuous Complex Queries In Structured Overlay Networks

Current peer-to-peer (P2P) has a rapid development as distributed computing model. Compared to the traditional Client/Server model, each node serves as user and service providers at the same time which makes it very suitable for large-scale network applications with strong expansion and fault tolerance. Flow data model in a variety of applications has been extensive researched, including network monitoring, financial data analysis, sensor networks and so on. For these applications, compared to traditional databases, databases based on Peer-to-Peer have a distinct advantage more suitable for distributed data processing flow. The basic structure of particular DHT using peer-to-peer topology which limits it that can only support simple exact match query, so how structured peer-to-peer supporting for the complexity of distributed query is a very meaningful question. This article focuses on how to achieve continuous queries over structured peer-to-peer network with update data frequently, which can provide sorted queries just like web search engine functions, which returns the Top-k results of the most requested to the user.In this paper's environment, the data is organized into relational data model tuples which are inserted in the system from different nodes continuously. As a result of the use of structured peer-to-peer DHT, this particular topology, that is, mapping a single keyword to find the node, thus limiting its ability to carry out complex queries. How to combine different data from the network nodes associated stream data processing, while minimizing the resulting flow is a challenging task. In this paper, two kinds of continuous Top-k join query algo- rithms were proceeded, For ACJoin algorithm, all the data received will trigger queries. Before join queries, Algorithm PCJoin first compute the Rank-f sort function's value and then compared with the Top-k results to decide whether this data is useful for Top-k results which reduces the resource consumption and network traffic. Finally, a large number of simulation experiments carried out performance analysis, experimental results show that the algorithm to ensure availability for Top-k query results and at the same time reducing a large amount of network traffic.