隨著資訊科技的快速發展，所需儲存的資料量越多，資料的儲存與管理扮演很重要的角色。因為人們發現單一位置的資料存放並不是一個好的做法，可能會因資料庫系統毀損、硬體故障、甚至遇到天災等因素而遺失資料。此外，對於資料毀損的企業們，必須付出許多的時間與金錢來重建。若想要確保重要資料的安全性，則資料備份與還原的重要性便顯得越來越被人們所重視。 本論文提出一個支援多分割區系統之快速備份與還原演算法設計，提供穩定且有效率的資料備份與還原功能，以及同時支援Windows、Mac、Linux平台上主流的4種檔案系統格式，即NTFS、Extended FAT、HFS Plus、Ext2。最後實驗結果顯示，此種快速演算法(Intelligent Algorithms)的備份與還原速度皆比完全演算法(Full Algorithms)更有效率，其改善後的速度可達2倍。因此，該論文提出的快速演算法具有良好的效能與可擴充性，並滿足企業大量地資料備份與還原之需求。 With the rapid development of information technology, the amount of data storage is increased, and then data storage and management play an important role. People found that it is not a good way to storage data in a single location because data may be lost due to database corruption, hardware failures, and even natural disasters. Moreover, when the enterprises encounter the accidental loss of data, they must spend more time and money to recover these lost data. If we would like to ensure the data security, data backup and recovery techniques are attached great importance. In this thesis, the intelligent backup and recovery algorithms design for multi-partition system are proposed. Our algorithms provide the steady and efficient data backup and recovery functions, and support four file system formats such as NTFS, Extended FAT, HFS Plus and Ext2. Finally, the experiment results show that backup speed and recovery speed of our intelligent algorithms are more efficiently than full algorithms, and the improved speed up to two times. Thus, our designed algorithms have good performance and scalability, and meet the requirement of enterprise massively data backup and recovery.
