CLOUD QUERY PROCESSING ANALYSIS:
ENCRYPTION AND DECRYPTION
Zainalabideen Ali
Faculty of Information Science and Technology, National University of Malaysia, Malaysia.
E-mail: gp06356@siswa.ukm.edu.my
Azana Hazah Binti Mohd Aman
Faculty of Information Science and Technology, National University of Malaysia, Malaysia.
E-mail: azana@ukm.edu.my
Rosilah Hassan
Associate Professor, Faculty of Information Science and Technology, National University of
Malaysia. Selangor, Malaysia.
E-mail: rosilah@ukm.edu.my
Recepción: 26/07/2019 Aceptación: 18/09/2019 Publicación: 06/11/2019
Citación sugerida:
Ali, Z., Aman, A.H.B.M. y Hassan, R. (2019). Cloud query processing analysis: encryption
and decryption. 3C Tecnología. Glosas de innovación aplicadas a la pyme. Edición Especial, Noviembre
2019, 65-75. doi: http://dx.doi.org/10.17993/3ctecno.2019.specialissue3.65-75
Suggested citation:
Ali, Z., Aman, A.H.B.M. & Hassan, R. (2019). Cloud query processing analysis: encryption
and decryption. 3C Tecnología. Glosas de innovación aplicadas a la pyme. Special Issue, November
2019, 65-75. doi: http://dx.doi.org/10.17993/3ctecno.2019.specialissue3.65-75
66
67
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254–4143
ABSTRACT
The usage of clouds to provide data query services is becoming an attractive
solution to services that demand scalability and cost minimization. Despite the huge
advantages, cloud consumers require their condential or delicate data to be safe
and secured, because violating their private data would be a great concern that is not
tolerable. Therefore, corporations who use the cloud services especially the database-
as-a-service have tended to encrypt sensitive data for security and condentiality
purpose. Encrypting data would facilitate protecting private information from any
violation by the service provider. Several studies have proposed AES and RSA
encryption methods. Yet, each encryption method provides a specic level of security
which comes with an opposite level of eciency. Therefore, this work will examine
the performance for both encryption methods. The environment setup consists of
Microsoft SQL Server as cloud database simulation, and a Visual Studio platform to
simulate the local processing of queries. The performance evaluation parameters are
time consumption for encryption and decryption. Overall, the results appeared are
not signicant as each method has its own benets.
KEYWORDS
RSA, AES, Processing Time, Cloud Query Processing.
66
67
Edición Especial Special Issue Noviembre 2019
DOI: http://dx.doi.org/10.17993/3ctecno.2019.specialissue3.65-75
1. INTRODUCTION
Cloud computing is a term that was introduced in 2006 after the introduction of
Amazon’s Elastic Compute Cloud (EC2) (Singh, 2015) dened a cloud computing
model as a model that facilitates easy and on-demand network access to a shared pool
of congurable computing services and resources. These resources include networks
(Aman, 2019), servers, storage, apps, and services. Modak, Chaudhary, Paygude, and
Ldate (2018) noted that there are three components of cloud computing, which are
data centers, the client computer, and distributed servers. A company that intends to
create a database would prefer to rent an existing one provided by the cloud rather
than initiating it from scratch, where there is a third party that can manage and
controls these services (i.e. a virtual world) (Singh, Rishiwal, & Kumar, 2018).
Despite cloud computing has huge advantages. Cloud consumers do not wish to
place their condential or delicate data, such as health records (Ibrahim, 2019),
emails, and govern-mental important les; because violating their private data would
be a great concern that is not tolerable. Encrypting data would facilitate protecting
private information from any violation by the service provider (Vurukonda & Rao,
2016). The daily usage of query processing over an encrypted data would require
a long time where the query needs to be encrypted, and the retrieved data will
be decrypted to be shown for the employees. Apparently, the eciency would be
signicantly impacted. Recent eorts in the cloud query processing have shown some
attempts to improve the eciency of processing encrypted queries (Amini, 2018).
However, there are still some concerns regarding the privacy in which the encryption
type would be examined (Albadri & Sulaiman, 2016; Ghaleb, Shukur, Sulaiman, &
Mobidin, 2018).
This paper consists of four (4) sections. Section I introduces an overview of cloud
computing, Section II methods of evaluation. Section III presents the results and
discussion. Lastly, section V concludes the paper with a summary of the ndings and
recommended future work.
68
69
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254–4143
2. RESEARCH METHODOLOGY
The aim objective of this paper is to analyze the performance of two encryption
methods namely: RSA and AES. SQL Server is used as a cloud database simulation
and have been simulated the local processing of queries by using a Visual Studio
platform. As shown in Graphic 1, the architecture of the proposed method contains
three key components namely the local side, Encryption and Decryption methods
(AES and RSA), and the cloud side. The queries include three types which are Insert
data, retrieve data, and drop data.
Graphic 1. Model of The Proposed Method.
The dataset is stored in Microsoft SQL Server Management Studio, for the
experiments, a set of queries is prepared. This set contains three types of queries
including ‘Add’, ‘Delete’ and ‘Select’. Each type of query gives dierent results from
where security and time consumption for each method. Each type of query is run 50
times with dierent data. The database name is Student with two tables represents
68
69
Edición Especial Special Issue Noviembre 2019
DOI: http://dx.doi.org/10.17993/3ctecno.2019.specialissue3.65-75
RSA and AES table. The data type is using the non-variable character (nvarchar),
since using nvarchar rather than varchar, avoid encoding conversions which consume
time every time it reads from or writes to the database. Conversions take time and are
prone to errors. The nvarchar size is depending on the key length for each method,
200 for AES and 2000 for RSA.
The computer which has been utilized having intel core i7-7500U CPU @2.70GHz
2.90GHz processor with 8GB RAM, system type 64-bit, Windows 10 Pro and
computer type is LENOVO Ideapad 310. The AES has been implemented using a
C# library called AesManaged, while the RSA has been implemented a C# library
called RSACryptoServiceProvider. The interface of the proposed method enables
the user to type a query from the predened ones. Then, either by using AES or
RSA encryption methods the interface will encrypt the query and attempt to execute
it, meanwhile, decrypt the retrieved results to present it to the user. With regards the
encryption and decryption, for RSA, the asymmetric key was 2048-bit key, while for
AES, symmetric key generation using the 128-bit key.
3. RESULTS AND DISCUSSION
The results of applying the encryption time including AES encryption method, and
RSA encryption method. Graphic 2, Graphic 3 and Graphic 4 shows the encryption
is being performed for ‘Add’, ‘Select’, and ‘Delete’ queries respectively.
While the results of applying the dencryption time for AES and RSA method is
shown in Graphic 5, Graphic 6 and Graphic 7 being performed for ‘Add’, ‘Select’,
and ‘Delete’ queries respectively.
70
71
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254–4143
Graphic 2. Encryption Time for Add.
Graphic 3. Encryption Time for Select.
Graphic 4. Encryption Time for Delete.
72
73
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254–4143
Obviously, the “Add” queries have taken the longest time to complete, given that
Add process requires inserting new records which consume more time, compared to
displaying existing record as in the “Select” query, or deleting existing record as in
the “Delete” query. In general, the performance of the proposed study tends to be
competitive between AES and RSA in terms of eciency.
The results are not signicant, as the percentage of improvement for each query
types are calculated. It has appeared 23% for Add command, 2% for Select com-
mand, and 2% for Delete command between the two encryption methods. The RSA
with 2048 key length has given slightly better results in terms of the processing time:
“Add”, “Select”, and “Delete” when the data is set to nvarchar 2000. This might
be because of the selection of key size; data type and data size play an important
metric criterion in choosing the encryption methods. However, it is recommended,
methods should be chosen based on types of cloud applications and services, as well
as possible attacks and threats.
4. ACKNOWLEDGEMENT
The authors are grateful to Faculty of Information Science and Technology, National
University of Malaysia. This research is also funded by research grant DIP-2018-
040.
5. CONCLUSIONS
The potential enhancement that could be used in future studies are using another
encryption method may improve encryption time and encryption security, suggest
a combination of multiple encryption methods in order to enhance a part of the
encryption time and enhance the security of some part, and using real-time dataset
could provide some advantage regarding the tactical issue.
72
73
Edición Especial Special Issue Noviembre 2019
DOI: http://dx.doi.org/10.17993/3ctecno.2019.specialissue3.65-75
REFERENCES
Albadri, H., & Sulaiman, R. (2016). A Classication Method for Identifying
Condential Data to Enhance Eciency of Query Processing over Cloud. Journal
of Theoretical and Applied Information Technology, 93 (2), 412-20.
Aman, A.H.M., Hassan, R., Hashim, A.H.A., and Ramli, H.A.M. (2019).
Investigation of Internet of Things Handover Process for Information Centric
Networking and Proxy Mobile Internet Protocol. In 3rd International Multi-Topic
Conference on Engineering and Science 2019.
Amini, R., Sulaiman, R. & Abd Rahman Kurais, A. H. (2018). CryptoROS:
A secure communication architecture for ROS-based applications. International
Journal of Advanced Computer Science and Applications, 9(10), 189-194.
Ibrahim, R., Aman, A. H. M., Nur, A. M., & Aljunid, S. M. (2019). Cost Centric
Data Mining for Radiology Procedures at Teaching Hospital in Malaysia. In 3rd
International Multi-Topic Conference on Engineering and Science 2019.
Ghaleb, H. S. M., Shukur, Z., Sulaiman, R. & Mobidin, H. S. (2018).
Implementation of AES algorithm in QGIS software. Proceedings of the 2017 6th
International Conference on Electrical Engineering and Informatics: Sustainable Society Through
Digital Innovation, ICEEI 2017. Institute of Electrical and Electronics Engineers Inc., Vol.
2017-November, 1-6.
Gupta, P., Verma, D. K. & Singh, A. K. (2018). Improving RSA Algorithm
Using Multi-Threading Model for Outsourced Data Security in Cloud Storage.
Proceedings of the 8th International Conference Conuence 2018 on Cloud Computing, Data
Science and Engineering, Conuence 2018, 163-69.
Modak, A., Chaudhary, S.D., Paygude, P.S., & Ldate, S.R. (2018). Techniques
to Secure Data on Cloud: Docker Swarm or Kubernetes? 2018 Second International
Conference on Inventive Communication and Computational Technologies (ICICCT), 7-12.
74
75
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254–4143
Singh, K. P., Rishiwal, V., & Kumar, P. (2018). Classication of Data to Enhance
Data Security in Cloud Computing. Proceedings-2018 3rd International Conference On
Internet of Things: Smart Innovation and Usages, IoT-SIU 2018, 1-5. doi: https://doi.
org/10.1109/IoT-SIU.2018.8519934
Singh, M. (2015). Study on Cloud Computing and Cloud Database. International
Conference on Computing, Communication and Automation, ICCCA 2015, 708-13.
Vurukonda, N., & Rao, B. T. (2016). A Study on Data Storage Security Issues
in Cloud Computing. Procedia Computer Science, 92, 128-35. doi: https://doi.
org/10.1016/j.procs.2016.07.335
