Speedup for Cryptography on CUDA Heterogeneous Architecture

Cryptography is an indispensable tool for protecting information in computer systems The Advanced Encryption Standard (AES) is one of the cryptography algorithms, which a well-known symmetric key algorithm, that utilizes the same secret key to encrypt and decrypt a message. It defeats that constraint of the more modest enter measure of the Data Encryption Standard (DES) toward using a greater and variable-length way. Moreover, It also determines those moderate preparing speed for triple DES (3DES), also uses more level assets of resources, therefore, it is preferred as the encryption and decryption standard by the U.S. government. AES is now being used worldwide for encrypting advanced information, including financial, telecommunications, and legislature data. In this paper, the AES scheme is implemented on Graphics Processing Unit (GPU) by parallelizing each step in the encryption algorithm of the AES using Compute Unified Device Architecture (CUDA). This allows enhancing the performance of the encryption algorithm execution to achieve a speedup factor of 61.36x on the multicore heterogeneous system.