Implementation and Performance Analysis of Modern Encryption Techniques in Cloud Computing

Abstract

Cloud computing offers scalable and cost-effective solutions for data storage and processing, enabling widespread adoption by IT service providers. However, data security remains a major concern. This paper evaluates several symmetric encryption algorithms, including AES, DES, 3DES, RC2, RC4, and Blowfish, within a cloud environment and proposes a selective encryption approach that dynamically chooses the optimal algorithm based on runtime performance. Performance is assessed using statistical and operational criteria, including randomness (via the NIST test suite), rejection rate, avalanche effect, execution time, and visual analysis. Experiments are conducted on Amazon EC2, with encryption implemented in Java and statistical tests in C#. Results show that no single algorithm performs best in all scenarios. RC4 is computationally efficient but weak in randomness and rejection rate. Blowfish and the Selective method offer strong randomness and security but with higher decryption times. AES provides the best avalanche effect. For high-security applications, Selective, Blowfish, AES, and 3DES are preferred, while RC4 and RC2 are suitable when speed is critical.