Study on Optimizing Block Ciphers (AES, CHAM) on Graphic Processing Units

With the development of IoT and cloud technologies, it is necessary to efficiently encrypt a large amount of data. GPUs (Graphic Processing Units) are considered as a proper candidate for a crypto accelerator in these situations. In this paper, we study how to optimize AES, a representative standard block cipher algorithm, and CHAM, a recently proposed lightweight block cipher algorithm, in the GPU environment. As a result of applying optimization techniques that make full use of algorithmic properties and computational characteristics in the GPU environment, we could achieve performance improvement up to a dozen times than classical CPU implementation on a 64-bit operating system on AMD RYZEN 5 3600 CPU and NVIDIA RTX 2070 SUPER GPU.