隨著網路的發達及企業的全球經營方式,檔案資料大部分儲存在不同的區域網路,所以檔案管理系統將面臨檔案的存取效率以及確保取得的檔案內容一致的問題。針對檔案分散儲存與檔案存取的問題,一般解決方法是將檔案集中儲存管理或檔案集中管理後,透過備份或複製檔案系統與資料到其他區域網路。雖然檔案集中管理可以確保檔案一致性,卻造成外部網路的使用者須付出更多時間與網路頻寬來存取檔案資料。而採取異地備援或檔案複製的方法將造成不必要的資料儲存空間與頻寬浪費,且此方法須搭配複雜的資料同步機制。木論文提出一個透過階層式邏輯檔案索引與快取的設計建立於分散式文件管理系統中的檔案存取機制,在現有網路頻寬與架構下,提供更有效率的檔案傳輸與文件管理。
More and more documents are stored dispersedly in geographically distributed branches. People often have to work over networks slower than LANs. How to keep the consistency of document states, and how to guarantee the performance of document accesses are the major problems for most enterprise, especially for the administrator of information management departments. To solve such problems, traditional mechanisms suggest a central sewer with a large storage containing all documents in it to provide the document access service. To satisfy enterprises' needs, this paper is proposed to provide an efficient and high performance distributed document access system for geographically separated users. With our proposed method, enterprises can upgrade the document management efficiency based on currently existent network topology and architecture. The high performance distributed document management system (DMS) is composed of two major modules, and they are the logic document index management module, and the file sewer module. Logic document index management module serves the adaptive distributed document access mechanism and meta-data management which is independent of physical document storages, and guarantees the consistency of document states. The file sewer module manages the mapping between hierarchical logic indices and physical storage directories. Besides, it supplies cache management functions with the same performance as local document access no matter where users are physically located.
為了持續優化網站功能與使用者體驗,本網站將Cookies分析技術用於網站營運、分析和個人化服務之目的。
若您繼續瀏覽本網站,即表示您同意本網站使用Cookies。