Research On Auditing Protocol For Outsourced Data Supporting Dynamic Operations In Cloud

Cloud storage is a significant service in cloud computing,which has great application and practical significance in real life.Storing data in cloud can not only reduce local costs in clients'side but also provide endless storage resources to these clients.However,on the one hand,since cloud data are not always correct,data updating is in necessary.Although data owners can modify the cloud data,with limited resources,they can not identify and modify all errors/mistakes in cloud data.What's more,since data owners may have the wrong subjective consciousness,modifications maybe not convincing enough.On the other hand,since cloud service providers are often sim-trusted,auditing protocols are in necessary.However,existing auditing protocols supporting data dynamic operations are with low efficiency.Some of them cost too much,and others are inefficient to support dynamic operations.Meanwhile,for the timestamp used in auditing is generated by data owners,and auditors are employed by data owners and serve them,cloud service providers may suffer from collusion attacks from data owners and the auditor.Therefore,in this paper we proposed the corresponding solutions to the mentioned problems,which are shown as follows:1)A dynamic operations supporting protocol with assistance of multi-assistants in cloudIn this paper,a protocol employing multiple assistants is proposed to support dynamic operations on outsourced data in cloud.We introduce the concept of "multi-assistant" in the protocol,which is actually a group of special cloud clients equipped with professional knowledge.Assistants are responsible for discovering the inappropriateness in cloud data and discussing with each other to obtain the appropriate modification suggestion,while the data owner execute the modification given above.As a result,the burden on data owners to find and correct errors in cloud data will be reduced,and the accuracy of modifications will be improved theoretically.2)An efficient public auditing protocol based on a new dynamic structureIn this paper,a novel dynamic structure consisting of doubly linked info table and location array is designed.There is a proper mapping between index numbers and physical locations of data blocks by employing the structure,so that the dynamic operations and batch auditing are more convenient and efficient.Further more,we design an auditing protocol for cloud data based on the dynamic structure.The auditing protocol consists of two parts:global and sampling verification,which can resist the collusion attack mentioned above and implement efficient data auditing.Note that sufficient numerical analysis and experimental resultsindicate that the computational and communication cost of the auditing protocol is relatively low.