Existing DHT algorithm as a distributed hash function, only the search for accurate, if you want to support the current Web search engine with multi-keyword search function, but also the introduction of new methods. The main reason is that DHT works.
DHT-based P2P system is compatible with the hash function under the precise positioning of key words and found objects. Hash function always try to ensure the hash value generated uniform random distribution, the result is high similarity of the two elements, but not exactly the same object is generated completely different hash value, stored in a completely random two nodes . Thus, DHT can provide an exact match query, but the semantic is very difficult.
DHT is currently conducted on the basis of resource management with semantic technology research also very small. Since the exact keywords DHT mapping is such that can not and the fields of information retrieval research results combined, prevented DHT-based P2P system large-scale application.
P2P Discovery of the most important research results should be based on small world theory, unstructured discovery algorithm and structured DHT-based discovery algorithm. DHT and its findings, especially the organization of technology resources and find it provides a new approach.
With the development of the practical application of P2P systems, the physical network, routing some of the factors that influence began to affect the efficiency of P2P discovery algorithm. On the one hand, the actual network nodes reflect the great differences between that heterogeneity. As client / server model in the field of Internet and distributed applications over ten years and a lot of the popular types of electronic equipment, such as mobile computers, mobile phone or PDA. These devices in computing power, storage space and battery capacity, very different. In addition, the actual network routers and switches are divided into different autonomous regions, reflecting strict hierarchy.
On the other hand, the extent of fluctuations seriously affect the network discovery algorithm efficiency. Network fluctuations (Churn, fluctuation of network), including node join, quit, failure, migration, concurrent accession process, network segmentation. DHT discovery algorithms such as Chord, CAN, Koorde fluctuations are taken into account worst case network design and Its Implementation. As the degree of each node to maintain the minimum possible, so that members need to respond to changes in the maintenance of relations can be relatively small, so you can quickly recover the impact of network fluctuations. However, only a small number of each node routing state of the high cost of delay discovery algorithm, because every search need to contact multiple nodes, in a stable network, this way of thinking is unnecessary.
Meanwhile, as a resource organization and discovery technology necessary to support complex queries, such as keywords, content querying. Although the field of information retrieval and data mining provides a wealth of sophisticated semantic search technology, due to the characteristics of DHT exact keywords maps hinders DHT complex queries in the application.
Recommand Link:
Expert GAMES ArcadeAvi to 3gpArticles about Access ControlPowerVideoMaker for PowerPoint 2000Youtube video formatLohan WMV to DVDYoutube FLV to PS3 TodaySwift DVD to MPEG4 ASF MPEGSwift SVCD VOB RIPPER PlatinumDVD To iPod + iPod Converter Suite MACM4A MP2 to BlackBerryLists Food And DrinkBRIEF Text Or Document EditorsFlac to wavConverter avi to 3gpMobile PHONE BurnerYoutube To Flash Now
