一個同儕式網路環境中基於完善差異圖與集合之高效率查詢機制

Abstract

在本篇論文中，我們提出一個嶄新並且有效率的查詢機制用於多層式非結構化同儕網路。此外我們的查詢機制不但具備可靠性，例如若某些資源存在系統中，我們能夠更有效率地送出廣播查詢訊息並且能夠成功地找到所有檔案資源; 而且具備延展性，例如網路中查詢訊息的流量能夠被超級同儕所限制住。此外在我們多跳數多階層的架構下，可以藉由更強大的超級同儕來連結其他的完善差異網路。就我們調查過多數研究所知，幾乎沒有研究針對非結構化同儕網路來提出同時兼具可靠性和延展性的查詢機制。我們採用多跳指標復製造(Multi-hop Index Replication)和完善差異圖傳遞演算法，可使得每一個超級同儕不會收到額外的廣播查詢訊息;此外藉由查詢Multi-hop Index，使得每一個超級同儕得以判斷檔案資源來自於何方，更進一步有效的減少網路流量。

In this thesis, we propose an novel and efficient search approach for multi-layer unstructured P2P systems, and show that it is not only reliable, i.e., if any content is in the system, it successfully locates all files and the search should find the content with reasonable guarantees, but also scalable, i.e., the network traffic generated by queries will be limited by super-peer. In addition, we can use more powerful super-peer to connect to other Perfect Difference Network (PDN). To the best of our knowledge, there are few researches focused on the reliable and scalable search mechanism for unstructured P2P systems. The broadcasting performance of the P2P system is enhanced through the use of a Multi-hop Index Replication with Perfect Difference Graph (PDG) forwarding algorithm, which make certain that each super-peer receives just one copy of the broadcast message. Furthermore, by using the Multi-hop Index, super-peer has extra information to know what the files it queries is available or unavailable files, and demonstrate their effectiveness and efficiency through the simulation.