Security and Performance Assessment of Encryption Techniques for Cloud Platforms

Abstract

Cloud computing platforms are becoming more popular for storage and processing of sensitive data, because of their scalability and flexibility, however the shared and distributed nature of cloud computing presents significant security dangers. Encryption is also a key element to keeping cloud data safe, but encryption operations will also add computational overhead to the system, affecting performance. This work brings an orderly evaluation of typically used encryption techniques in cloud platforms from both perspective of security effectiveness and performance efficiency. Symmetric, asymmetric and hybrid encryption approaches are examined under conditions that are relevant for cloud computing based on different criteria, e.g. encryption, decryption time, and computational overhead. Experimental results indicate that symmetric encryption performs the best on both encryption and decryption time (118 ms and 104 ms, respectively), which is applicable to bulk data protection, and asymmetric encryption causes high processing overhead (362 ms encryption time and 341 ms decryption time). Hybrid encryption offers a balanced performance where the processing times are moderate (176ms encryption and 158ms decryption) but more secure due to secure key management. The results point to some very clear tradeoffs between security strength and efficiency of performance and suggest that hybrid encryption represents a workable tradeoff for multi-user and scalable cloud environments. These findings constitute some good information in choosing encryption strategies that promote secure and efficient operations of clouds.