Pages

Tuesday 30 October 2012

Cooperative Provable Data Possession for Integrity Verification in Multi-Cloud Storage


ABSTRACT:

Provable data possession (PDP) is a technique for ensuring the integrity 

of data in storage outsourcing. Here, we address the construction of an 

efficient PDP scheme for distributed cloud storage to support the scalability of 

service and data migration, in which we consider the existence of multiple 

cloud service providers to cooperatively store and maintain the clients’ data. 

We present a cooperative PDP (CPDP) scheme based on homomorphic 

verifiable response and hash index hierarchy. We prove the security of our 

scheme based on multi-prover zero-knowledge proof system, which can satisfy 

completeness, knowledge soundness, and zero-knowledge properties. In 

addition, we articulate performance optimization mechanisms for our scheme, 

and in particular present an efficient method for selecting optimal parameter 

values to minimize the computation costs of clients and storage service 

providers. Our experiments show that our solution introduces lower 

computation and communication overheads in comparison with non-

cooperative approaches.


CONCLUSION:                       
                       
                 we presented the construction of an efficient PDP scheme for 

distributed cloud storage. Based on homomorphic verifiable response and hash 

index hierarchy, we have proposed a cooperative PDP scheme to support 

dynamic scalability on multiple storage servers. We also showed that our 

scheme provided all security properties required by zeroknowledge interactive 

proof system, so that it can resist various attacks even if it is deployed as a 

public audit service in clouds. Furthermore, we optimized the probabilistic 

query and periodic verification to improve the audit performance. Our 

experiments clearly demonstrated that our approaches only introduce a small 

amount of computation and communication overheads. Therefore, our solution 

can be treated as a new candidate for data integrity verification in 

outsourcing data storage systems. As part of future work, we would extend 

our work to explore more effective CPDP constructions. Finally, it is still a 

challenging problem for the generation of tags with the length irrelevant to 

the size of data blocks. We would explore such a issue to provide the support 

of variable-length block verification.

1 comment: