Proposing DEM Model for Improving the Security of Data in Cloud Environment

Ankita Kaushik, Amit KumarInternational Journal of Computer Applications 2017 · DOI

Abstract

Cloud computing is one of the most rapidly developing fields in computer science, offering services ranging from software and infrastructure to Database as a Service (DBaaS). However, security concerns regarding data privacy remain a significant barrier to adoption. Users are often reluctant to store sensitive information on cloud servers due to a lack of trust. This paper introduces the Data Encryption Model (DEM), an innovative approach to data security that utilizes field-based encryption at the client side to ensure data remains confidential even within an untrusted cloud environment.

Introduction

As organizations increasingly migrate to the cloud, Database as a Service (DBaaS) has become a cornerstone of modern infrastructure. Despite its benefits, a significant "trust gap" exists between users and Cloud Service Providers (CSPs). Many users fear that their sensitive data could be exposed through insider threats at the CSP or external breaches of the cloud infrastructure.

Current security solutions often rely on the provider's internal measures, leaving the data vulnerable if those measures fail. There is a critical need for mechanisms that allow users to leverage cloud storage and processing power while keeping the data itself a "black box" to the provider. The Data Encryption Model (DEM) addresses this by moving the security layer from the server to the client.

The Data Encryption Model (DEM) Architecture

The proposed DEM architecture acts as a secure proxy between the user and the cloud database. By performing encryption and decryption on the client side, the model ensures that plaintext data never leaves the user's controlled environment.

The Data Encryption Model (DEM) architecture showing the flow from user to cloud server.

The key components of the DEM architecture include:

  • Query Parser & Rewriter: When a user submits a SQL query, the client-side parser identifies sensitive keywords. The rewriter then transforms the query into an encrypted format before transmission.
  • Metadata Handler: This component manages database schema information locally, preventing the cloud server from learning the underlying structure of the data.
  • Arithmetic Engine: Designed to facilitate operations over encrypted data, allowing for basic processing without full decryption.
  • Encryption/Decryption Interface: This core logic implements "field-based encryption." Instead of encrypting an entire database file, individual fields (e.g., name, salary) are transformed independently.

Encryption Logic

For demonstration purposes, the authors utilized a character-based transformation algorithm:

  • ENCR (Encryption): Iterates through each character of a string. If the ASCII value is even, it is multiplied by 12; if odd, it is multiplied by 13.
  • DECR (Decryption): Performs the inverse operation (dividing by 12 or 13) to restore the original data upon retrieval by the client.

Implementation and Results

The model was validated using a prototype built with a .NET front end and a SQL Server backend. The authors used a custom "Emp" (Employee) table containing fields such as Empid, Empname, Deptname, Empsal, and Empholiday.

Demonstration of the encryption process where input values are transformed into ciphertext before saving.

The prototype successfully demonstrated:

  1. Functional Validation: Data entered into standard web forms (e.g., Name: "Shiv") was transformed into unreadable ciphertext before being stored in the database.
  2. Secure Retrieval: The client could execute "SELECT" queries, retrieve encrypted blobs, and use the local decryption algorithm to display plaintext to authorized users.
  3. Transparent User Experience: The complexity of query translation and encryption remains hidden from the end-user, who interacts with standard interfaces.

Discussion and Future Work

The DEM provides a robust architectural framework based on a "Zero Trust" assumption regarding the cloud server. By maintaining field-level granularity, the system preserves the relational structure of the database while hiding its content, allowing for more flexible querying than full-disk encryption.

However, several limitations exist. The current encryption algorithm is a basic substitution-style cipher intended for proof-of-concept; it would not withstand sophisticated modern cryptanalysis. Additionally, moving processing tasks to the client side introduces computational overhead. Future research should focus on implementing more robust, attribute-based encryption methods and expanding the system's ability to handle complex database operations like joins and aggregations on encrypted data.

Conclusion

The Data Encryption Model offers a promising path toward bridging the trust gap in cloud computing. By empowering users with client-side control over their data's security, DEM ensures that sensitive information remains protected, fostering greater confidence in the adoption of cloud-based database services.

Summary updated

Abstract

Cloud computing, which is the most developing fields of computer science. Apart from many services of cloud computing like software, hardware, infrastructure, desktop as a service and database also available as a service. But along with these, issues regarding security of data or database are of most fear concerns. As cloud users are reluctant to save data on cloud servers, which can be hindrance to this super crucial field of computers. Now the need of the hour is to develop an understanding that storing data on cloud servers is safer and also develop some mechanisms to improve the security of data. New researches are coming up with innovative approaches to provide more safety to data. In this paper, a DEM [Data Encryption Model] Model is introduced with a new approach for data security mechanism. The field base encryption method is used in this model.
BibTeX
@article{kaushik2017proposing,
  author = {Ankita Kaushik and Amit Kumar},
  title = {{Proposing DEM Model for Improving the Security of Data in Cloud Environment}},
  journal = {International Journal of Computer Applications},
  year = {2017},
  doi = {10.5120/ijca2017913422},
}
Figure