See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/277905691

Security and Verification of Data in Multi-Cloud Storage with Provable Data

Possession

Article in International Journal of Computer Applications · May 2015

DOI: 10.5120/20549-2924

CITATIONS READS

2 99

2 authors, including:

Ramesh M. Kagalkar
K.L.E. College of Engineering & Technology
57 PUBLICATIONS 159 CITATIONS

SEE PROFILE

All content following this page was uploaded by Ramesh M. Kagalkar on 22 January 2018.

The user has requested enhancement of the downloaded file.

 International Journal of Computer Applications (0975 – 8887)
Volume 117 – No. 5, May 2015

Security and Verification of Data in Multi-Cloud Storage

with Provable Data Possession
Vrushali K Gaikwad Ramesh Kagalkar
PG Student Asst.Proffessor
Computer Engineering Department, Computer Engineering Department,
DYPSOET Pune-43(India) DYPSOET Pune-43(India)

ABSTRACT the benefit of maintaining their reputation or for saving

Data integrity verification is one of the biggest security issue money and storage space the service provider might neglect to
in cloud. To check integrity of data Provable data possession keep or deliberately delete rarely accessed data files which
is one of the method available. In this paper, we have created belong to an ordinary client. Thus, Security of the data stored
an efficient PDP method for distributed cloud storage, in over cloud is necessary for cloud service providers. Different
which we multiple cloud service providers are maintaining techniques like Provable data possessions [3] (or proofs of
and storing client’s data in cooperative way. This retrievability [4]) are important for a storage provider to prove
cooperatively working PDP method is based on indexing the integrity and ownership of clients’ data without
hierarchy & homomorphic variable response method. The downloading it. This property of checking proof without
security of our scheme is based on trusted third party auditor downloading makes large-size files and folders to check
and structure of zero-awareness proof , which can fulfill whether these data have been tampered with or deleted
reliability of awareness, completeness, and properties. without any need of downloading the data. This leads to
replace traditional hash and signature functions in data
Keywords storageoutsourcing.
Multiple Cloud, Cooperative Provable Data Possession, Some recently proposed schemes like Scalable PDP [5] and
Homomorphic Variable Reply, Zero awareness, Cloud Dynamic PDP [6][7] mainly focus on PDP issues at untrusted
Storage Security. servers in a single cloud storage provider. So they are not
suitable for a multi-cloud environment.
1. INTRODUCTION
To store data on cloud is one of the services offered by cloud 2. LITERATURE SURVEY
computing. Therefore, instead of storing data on local server , In this section we are presenting the different methods which
they can store it on the cloud service provider’s storage. are previously used for Provable Data Possesion
Cloud storage gives the facility for users to store their data techinique.We discuss some limitations and advantages of
without consideration of hardware and software management, these systems.
which gives advantages like capability of storing unlimited
data and ability to access data anywhere[1]. Cloud computing Researchers have proposed two basic approaches to verify
can integrate multiple cloud services together. Thus it availability and integrity of outsourced data in cloud storages,
provides high interoperability environment. Thus this called Provable Data Possession (PDP) [2] and Proofs of
distributed cloud environment is known as multi-Cloud. Retrievability (POR) [3].They have addressed the problem in
distributed cloud environments of provable data possession as
Cloud computing is differs from other information technology per following aspects: high security, transparent verification,
in three ways by, and high performance. B.Sotomayor[1], they present
A] Resources Outsourced – This includes both hardware and OpenNebula, an open source virtual infrastructure manager
software. On-site file server can provide a source for file that can be used to deploy virtualized services on both a local
handling, data storage, and information backup. pool of resources and on external IaaS cloud. Ateniese et al.
[2] proposed the PDP model for ensuring possession of
B] Pay-as-you-can –It require a basic starting fee followed by untrusted storages. They also proposed a publicly verifiable
a monthly usage charge. User need to pay charge based on version, which allows anyone, to challenge the server for data
cloud time consumption and additional software features. possession. This property greatly extended application areas
C] On-demand facility – In cloud computing, user can pay for of PDP protocol due to the separation of data owners and the
what they use. users. However, these schemes are insecure against replay
attacks in dynamic scenarios because of the dependencies on
In multi-cloud environment by using different interfaces the index of blocks. Moreover, they do not fit for multi-cloud
clients can access resources. Web services by using virtual storage due to the loss of homomorphism property in the
infrastructure management [2] is example it. Various tools verification process.
and technologies are available for multiple clouds such as
VMware, vSphere, and Platform VM Orchestrator[2]. These In order to support dynamic data operations, Ateniese et al.
tools help cloud providers for creating a platform for data developed a dynamic PDP solution called Scalable PDP [4].
storage. But, if such an important platform is susceptible to They proposed a lightweight PDP scheme based on
security attacks, these attacks may introduce irrevocable cryptographic hash function and symmetric key encryption,
losses to the clients. but the servers can deceive the owners by using previous
metadata or responses due to the lack of randomness in the
The biggest issues with cloud data storage is that of data challenges. The numbers of updates and challenges are
integrity verification at untrusted servers. For example, the limited and fixed in advance and users cannot perform block
cloud service provider (CSP) might suffers Byzantine insertions anywhere. Based on this work, Erway et al. [5]
failures. Such CSP’s hide the data errors from the clients for introduced two Dynamic PDP schemes with a hash function

10
 International Journal of Computer Applications (0975 – 8887)
Volume 117 – No. 5, May 2015

tree to realize O(log1) communication and computational The verification procedure is described as follows: Firstly, a
costs for a -block file. The basic scheme, called DPDP-I, client (data owner) uses the secret key to pre-process a file
retains the drawback of Scalable PDP, and in the 'blockless' which consists of a collection of n blocks, generates a set of
scheme, called DPDP-II. public verification information that is stored in TTP, transmits
the file and some verification tags to CSPs, and may delete its
Furthermore, these schemes are also not effective for a multi- local copy.
cloud environment because verification path of the challenge
block cannot be stored completely in a cloud. Juels and Then, by using a verification protocol, the clients can issue a
Kaliski [3] presented a POR scheme, which relies largely on challenge for one CSP to check the integrity and availability
preprocessing steps that the client conducts before sending a of outsourced data with respect to public information stored in
file to a CSP. Unfortunately, these operations prevent any TTP. Then neither assume that CSP is trust to guarantee the
efficient ex-tension for updating data. security of the stored data, nor assume that data owner has the
ability to collect the evidence of the CSP's fault after errors
3. VERIFICATION FRAMEWORK have been found. To achieve this goal, a TTP server is
OVERVIEW constructed as a core trust base on the cloud for the sake of
In this architecture, we consider the existence of multiple security.
CSPs to cooperatively store and maintain the clients' data.
Moreover, a PDP is used to verify the integrity and
availability of their stored data in all CSPs.

Fig. 1 System Architecture

We assume the TTP is reliable and independent through the 4. COOPERTIVE PROVABLE DATA
following functions to setup and maintain the CPDP
cryptosystem; to generate and store data owner's public key; POSSESSION SHCEME
and to store the public parameters used to execute the A PDP is a collection of two algorithms (Key Gen, Tag Gen)
veriffcation protocol in the CPDP scheme. The TTP is not and interactive proof system Proof.
directly involved in the CPDP scheme in order to reduce the
complexity of cryptosystem. Key Gen: It takes a security parameter as an input and returns
a secret key as output.

11
 International Journal of Computer Applications (0975 – 8887)
Volume 117 – No. 5, May 2015

Tag Gen: It takes a secret key, file and set of cloud storage 3. GenProof()
providers as input and returns triples. 4. CheckProof()
In this model, Symbol C is denote set component which used
Proof: It is a protocol of proof of data possession between the in this framework.
CSP's and veriffer.
C={pk,sk,m,Tm,F,chal,Σ,V}
Let H = Hk be a family of hash functions where k : {0 1}1
index by k ε K. Where,
This algorithm has a benefit in breaking the collision pk = public key,
resistance of H.Collision-Resistance H: In this a hash family
H(t, ε) collision resistant if no t-Time adversary has advantage sk = secret key,
atleast ε in breaking collision of H.First the KeyGen algorithm m = file block,
is run in this scheme to obtain the public or the private keyfor
users. Then TagGen is generated by the clients for the Tm = metadata,
outsourced data. F = collection of blocks ,
5. HOMOMORPHIC VERIFIABLE Chal = challenge,
RESPONSE FOR PDP Σ = verification metadata corresponding to blocks in F,
A homomorphism is a map f : P Q between two groups such
that f (g1+ g2 ) = f( g1 ) x f( g2 ) for all g1,g2 ε P, where + V= proof of possession
denotes the operation in P and x denotes the operation in Q. A PDP system can be constructed from a PDP scheme in two
Homomorphic verifiable response is the key technique of Phases, Setup and Challenge
CPDP because it not only reduces the communication
bandwidth, but also conceals the location of outsourced data A . Setup:-
in the distributed cloud storage environment.
Pk and sk runs by client i.e KeyGen
6. HASH INDEX HIERARCHY FOR TagBlok(): Input={pk,F, Σ=(Tm1,…….,Tmn)}
CPDP Output={F,Σ}
Three layers are used to illustrate the relationships among the
blocks for stored resources . They are as follows: B.Challenge:-
1.Express Layer: it shows representation of stored resources. Client sends challenge chal to Server S.
2.Service Layer : it offers and manages cloud storage and Genproof():-Input={pk,F,chal,Σ)
services and
Output={F,Σ}
3.Storage Layer : realizes data storage on physical devices
CheckProof():-Input={pk,sk,chal,V}
Output={success,failure}
Following polynomial-time algorithms will be used :-
◦ KeyGen(1k) → (pk,sk)
◦PrepareUpdate(sk,pk,F,info,Mc) →{e(F),e(info),e(M)}
◦PerformUpdate(pk,Fi−1,Mi−1,e(F),e(info),e(M))
→{Fi,Mi,M′c,PM′c}
◦ VerifyUpdate(sk,pk,F,info,Mc,M′c,PM′c) →{accept,reject}
◦ Challenge(sk,pk,Mc) → {c}
◦ Prove(pk,Fi,Mi,c) → {P}
◦ Verify(sk,pk,Mc,c,P) → {accept,reject}

8. SECURITY ANALYSIS:
For security purpose, cooperative scheme satisfies following
properties:
1) Collision resistant indexing: The indexing hierarchy in
CPDP scheme is collision resistant. If the client generates files
with the same file name and tries to store in multicloud,
Fig. 2 Hash Index Hierarchy collision because of name doesn’t occur there.

7. Mathematical Model 2) Public verification property: This Public verification

This proposed work uses below function as follows. property allows client as well as anyone other than client (data
owner) to challenge the cloud server for data integrity and
1. KeyGen() data ownership without the need for any secret information.
2. TagBlock()

12
 International Journal of Computer Applications (0975 – 8887)
Volume 117 – No. 5, May 2015

3) Zero-awareness property: Privacy of the data blocks stored on indexing, we have planned a cooperative scheme to
in multi-cloud and signature tags can be preserved by using support dynamic scalability using multiple storage servers.
this verification property. There are multiple cloud service providers for multiple clouds
as we are using multiple clouds.
4) Knowledge reliability verification: It is not possible to fool
the verifier easily to accept false statements. These structures Central Cloud Service Provider is used for minimizing the
can also oppose the tag forgery attacks, which help to avoid complexity as we want to store data block in each cloud, the
cheating the CSPs’ owner. This property is responsible for request has to go from each Cloud Service Provider. Thus,
avoiding tampering of the data or tag forgery, when collisions Cloud Service Providers manages requests. During uploading
tried. and downloading User has to answer the Security Question.
Security Questions and Answers are submitted by user during
9. MODULE INFORMATION the registration phase. So during Uploading/Downloading
Module1. Login and Registration operation If user is normal then he can answer that security
questions if he/she is intruder then he/she cannot answer that
In this we will develop the Login and Registration GUI for
questions. Thus, using this we can provide more Security.
Entities included in Project.
Also, we can use encryption algorithm [9] to provide the
Module2. Cloud Customer Security to uploaded data.
The Customer or User of the Cloud is one who has a large 12. REFERENCES
amount of data to be stored in multiple clouds and have the [1] Yan Zhu, Hongxin Hu, Gail-Joon Ahn, “Cooperative
permissions to use and access stored data. Before uploading Provable Data Possession for Integrity Verification in
process, User’s Data is converted into data blocks. That data Multi-Cloud Storage” IEEE, Mengyang Yu, Dec-2012
blocks are uploaded over multiple clouds in uploading
process. The TTP outlooks the data blocks Uploaded in multi [2] B. Sotomayor, R. S. Montero, I. M. Llorente, and I. T.
cloud. The user can also update the data uploaded over Foster, “Virtual infrastructure management in private and
multiple clouds. If the client wishes to download their files, hybrid Clouds,” IEEE Internet Computing, vol. 13, no. 5,
the data from multi cloud is integrated sequentially and pp. 14–22, 2009.
downloaded. [3] G. Ateniese, R. C. Burns, R. Curtmola, J. Herring, L.
Module3. Trusted Third Party Kissner, Z. N. J. Peterson, and D. X. Song, “Provable
data possession at untrusted stores,” in ACM Conference
Trusted Third Party (TTP) who is trusted to store verification on Computer and Communications Security, P. Ning, S.
parameters and offer public query services for these D. C. di Vimercati, and P. F. Syverson, Eds. ACM, 2007,
parameters. In this system the Trusted Third Party, outlook pp. 598–609.
the user data blocks and uploaded to the distributed cloud. In
distributed environment of cloud each cloud has user data [4] A.Juels and B. S. K. Jr., “Pors: proofs of retrievability for
blocks. If anybody tries to change the data stored over cloud Large files,” in ACM Conference on Computer and
Trusted Third Party gets alert of it that is again sent to client. Communications Security, P. Ning, S. D. C. di
Vimercati, and P. F. Syverson, Eds. ACM, 2007, pp.
Module4. Multi cloud storage 584–597.
In this system the each cloud admin will be having data [5] G. Ateniese, R. D. Pietro, L. V. Mancini, and G. Tsudik,
blocks stored over their cloud. Cloud computing has the “Scalable and efficient provable data possession,” in
ability to integrate multiple cloud services together to provide Proceedings of the 4th international conference on
high interoperability environment as it is established based on Security and privacy in Communication networks, Secure
open architectures and interfaces. Such distributed cloud Comm, 2008, pp. 1–10.
environment where multiple clouds are working cooperatively
is known as multi-Cloud. In this section, user uploads the data [6] Y. Zhu, H. Wang, Z. Hu, G.-J. Ahn, H. Hu, and S. S.
into multi cloud. Yau, “Dynamic audit services for integrity verification of
outsourced Storages in clouds,” in SAC, W. C. Chu, W.
10. RESULT E. Wong, M. J. Palakal, and C.-C. Hung, Eds. ACM,
In our PDP scheme, the client’s communication overhead is 2011, pp. 1550–1557.
not changed and the interaction among CSPs needs c−1 times
[7] C. C. Erway, A. K¨upc¸ ¨u, C. Papamanthou, and R.
constant-size communication overheads, where c is the
Tamassia, “Dynamic provable data possession,” in ACM
number of CSPs in multi clouds. Thus, the amount of
Conference on Computer and Communications Security,
communication overheads is not increased. Further, we
E. Al-Shaer, S. Jha, and A. D. Keromytis, Eds. ACM,
evaluated the performance of our PDP scheme in terms of
2009, pp. 213–222.
computational overhead. For comparison, our experiments
will executed as follows: a fixed-size file is used to generate [8] L. Fortnow, J. Rompel, and M. Sipser, “On the power of
the tags and prove data possession under the different number multiprover Interactive protocols,” in Theoretical
of sectors s. Then, the computational overheads of tag Computer Science, 1988, pp. 156–161.
generation are created. The results shows that the overheads
are reduced as the values of s are increased. [9] D. Boneh and M. Franklin, “Identity-based encryption
from the weil pairing,” in Advances in Cryptology
11. CONCLUSION AND FUTURE WORK (CRYPTO’2001), vol. 2139 of LNCS, 2001, pp. 213–
From research, we have presented an efficient method for 229.
security of data outsourced over multi-cloud. This research,
efficient method of PDP scheme is constructed for distributed
cloud storage. This scheme provided all security properties
required by zero knowledge interactive proof system. . Based

IJCATM : www.ijcaonline.org 13

View publication stats