| With the rapid development of cloud storage,more and more people choose to upload their data to cloud.Although the mode brings many benefits,one problem can not be neglected that cloud servers are untrusted.Cloud servers may keep data corruption from users due to various reasons,so data retrieved from the cloud is corrupted.Therefore,it has became a hot topic that how to check the integrity of data in cloud environment.Provable Data Possession(PDP)is one of important means of solving this problem,so this article will focus on the PDP.FlexList(Flexible Length-Based Authenticated Skip List,FlexList)based DPDP(Dynamic PDP)is one of latest researches in current that fully support dynamic update of data,however,cost is high for updating consecutive nodes.To solve this problem,We proposed optimized FlexList(O-FlexList)for updating of multiple consecutive nodes,including insertion,remove,modification and corresponding verification.Modification also supports random multiple nodes.Through analysis and test,the result shows that execution time and generated evidence length of O-FIexList have a great improvement for consecutive multiple nodes update,and the cost is consistent with the original algorithm for single node.We designed a distributed dynamic provable data possession(D-DPDP),and presented the formalized definition,interaction protocol among the entities and analysis of security and performance.In this model,there are three entities including users,a center server and storage servers.Server consists of a center server and some storage servers.The structure of server-side is transparent,and server-side provides service through the center server.Users don’t need to preprocess the data based on the structure of servers.This increases the system expansion ability,also the distributed authentic data structure and multiple copies improve the response time and data availability.Finally,we implemented and tested the D-DPDP prototype system,the result shows that the system can complete expected function. |
