Research On Key Technologies For Cloud Storage Data Integrity Auditing With Various Number Of Replica

The development of the Internet of Things,smart cities,digital economy and other emerging industrial models have brought huge benefits to people while also generating massive amounts of data,and traditional storage methods are no longer able to meet such huge data management needs.Therefore,individuals,enterprises and government users have migrated their data to the cloud.Cloud storage not only provides a large amount of storage space and low cost,but also provides more convenient access and more flexible capacity scalability.However,moving data to the cloud brings many benefits as well as security risks.These risks include,but are not limited to,data corruption,unauthorized access,data leakage,sensitive information compromise and privacy breach.Securing cloud storage data is not only critical for every ordinary user,but also for the economic and political security of an organization and even a country.Currently,how to secure cloud storage data technically has become an important research area in industry and academiaIn response to the various security risks mentioned above,key technologies for cloud storage security have also developed several sub-divisions such as ciphertext retrieval,data isolation,data integrity auditing,data privacy preservation,security authentication and access control.In this paper,we focus on four aspects of cloud storage data integrity auditing and propose a scheme for each of them.The first three are oriented to the big data storage scenario,and the last one focuses on single file scenario.In general,a multi-auditor,multi-cloud service provider data auditing approach is firstly designed based on blockchain,and then two approaches are successively proposed for the multi-replica multi-cloud storage provider scenario.The former approach mainly removes the huge overhead caused by computing and storing audit auxiliary information in the traditional approach,while the latter one optimizes the traffic overhead of communication between multiple servers in the network while ensuring the functionality and security of the former approach.The last approach in the article focuses on the small file scenario in cloud storage and designs an integrity detection method for a single image file.The details of the research are as follows:1.A smart contract-based method for auditing the integrity of cloud storage data is proposed for multi-auditor scenarios.The focus of the method is on how to take advantage of blockchain to solve the voting problem under multiple participants.Unlike some existing methods that use reputation as voting weights,the auditors in the proposed method vote equally and the membership for a specific audit task are constructed on an ad hoc basis,so no additional maintenance is required.This mechanism not only frees auditors from non-audit related trivialities,but also avoids the centrality problem due to excessive voting weights of certain groups.The challenges used to check data integrity are jointly generated by each of the participating auditors.As long as there is at least one honest auditor,collusion by any of the remaining parties will be detected.We rely on Ethereum to implement and deploy the solution as smart contracts.With the support of blockchain,the whole auditing process is open and transparent,where the data generated and the results obtained are immutable,which ensures the historical traceability of all audits.Theoretical analysis and experimental results show that the approach meets the designed security and functional goals,while featuring high efficiency and low gas overhead.2.A new collaborative cloud storage data integrity auditing model is proposed for multi-replica,multi-cloud storage provider scenarios.The approach aims to reduce the huge overhead caused by the traditional approach for computing and storing audit auxiliary information.In addition,the proposed approach provides unprecedented support for free chunking of data.This approach leverages the features of multi-replica and multi-cloud storage providers to combine the roles of auditor and storer in the traditional audit model into one,removing the role of a third-party auditor from the system.Another benefit is it removes the huge overhead associated with the computation and storage of auxiliary information.The proposed approach utilizes only hash functions in terms of computation,which results in negligible computational overhead compared to traditional methods based on bilinear pairings.Theoretical analysis and experimental results show that the proposed method provides high efficiency and flexibility with guaranteed security and can be used as a lightweight alternative to traditional cloud storage data integrity auditing methods in the scenario of multi-replica multi-cloud storage providers.3.To address the large network traffic overhead caused by broadcast communication among multiple parites in the previous approach,a block design-based cloud storage data integrity auditing method is proposed to further optimize the network configuration.In a collaborative audit with v cloud storage providers,the network traffic O(v2)required for the original broadcast communication is reduced to O(v√v),while eliminating the introduction of third-party auditors.In addition,the proposed approach uses a tag-free auditing model,i.e.,it eliminates the huge computation and storage overhead incurred by homomorphic verification tags.Outsourced files do not require any pre-processing for auditing before the uploading,and this improvement makes the approach more practical.Theoretical analysis and experimental results show that the proposed method is efficient and flexible while ensuring security? it is a practical data integrity auditing method for multi-replica and multi-cloud storage provider scenarios.4.Considering the access requirements of cloud storage users for independent files,an integrity checking method for single image files is proposed.Traditional cloud storage auditing is block-based for large-scale data,which is not flexible enough for small-scale file-based auditing.This method aims to solve this problem by using reversible data hiding to embed the original file digest into the image then upload it to the cloud storage,and after re-downloading the file,the file integrity is obtained by comparing the extracted and computed digests.This method makes the embedding operation less susceptible to malicious cloud storage providers by optimizing the embedding operation to reduce the corruption of the statistical features of the file.The method achieves the designed functional goals while better preserving multiple statistical features of the image,reducing the distortion of the image,and making the embedding operation less detectable.