A study of AES and RSA algorithms based on GPUs

Abstract

Combining the characteristics of GPUs and CPUs can greatly enhance the performance and bandwidth, though their design philosophies are inherently different. This research paper focuses on the GPU parallel computing technologies for encryption of AES and RSA algorithms. With the arrival of NVIDIA Technology’s Compute Unified Device Architecture (CUDA), the graphics processing unit (GPU) has shifted from graphics-capable devices to universal sequence streamline units. Cryptography is about the security techniques. The Advanced Encryption Standard (AES) is widely used in the file encryption, wireless security, SSL/TLS, etc. However, its execution efficiency can still be improved to provide better security. Similarly, improvements can also be made in serial implementations of the RSA algorithm. Parallel implementation of security algorithms has been shown to improve execution efficiency in comparison to serial implementation. This research paper focuses on parallel implementation of AES and RSA algorithms on GPU to enhance performance and energy efficiency in comparison to CPU