| Summary: | 碩士 === 國立中山大學 === 資訊工程學系研究所 === 102 === Cloud computing is increasingly popular today. Cloud services such as data-outsourcing services provide a growing number of users access to cloud storage for large quantities of data, and enterprises are turning to cloud storage for cost-effective remote backup. However, four limitations hinder the effectiveness of cloud storage: loss of availability, loss and corruption of data, loss of privacy, and vendor lock-in. DEPSKY is the only available system that overcomes all four limitations. Unfortunately, DEPSKY lacks an error detection mechanism and comes with heavy computing costs. Therefore, we propose a new data-outsourcing scheme overcoming not only the four limitations, but also the shortcomings of DEPSKY. In this thesis, we modify Nyberg’s accumulator and apply it to our three proposed error-detection methods. Moreover, we specially design a fast recovery method that is faster than DEPSKY and alternative methods.
|
|---|
