FlexDPDP: Esnek dinamik ispatlanabilir veri saklama

Cloud storage systems are becoming cheaper and more available. With the increase in popularity of the cloud storage systems both in industry and our personal lives people have started to care about the security of their data on the clouds. In this thesis, we develop and test a complete system for a server able to prove integrity of the client?s data without her downloading the whole data, and still letting the client interact with her data in a read/write manner. This system is called Dynamic Provable Data Possession by Erway et al. . We first show the FlexList: Flexible Length-Based Authenticated Skip List, a data structure optimized for secure cloud storage systems, and its differences from previous data structures. Then we demonstrate its utilization on the dynamic provable data possession system and we call the new scheme FlexDPDP. We further optimize the FlexDPDP scheme using parallelization techniques, and provide optimized algorithms to reduce the time complexity of the protocol. We provide an analysis on all of our proposals at the end of each chapter. We also deployed the optimized FlexDPDP scheme on large-scale network test-bed PlanetLab, demonstrating that FlexDPDP performs comparably to the most efficient static storage scheme (PDP), while providing dynamic data support. Finally, we demonstrate the efficiency of our proposed optimizations on multi-client scenarios according to real workloads based on real version control system traces.