CUDA Tutorial – Cryptanalysis of Classical Ciphers Using Modern GPUs and CUDA
Bulgarian Academy of Sciences, Institute of Mathematics and Informatics, Sofia, 1113, Bulgaria
arXiv:2103.13937 [cs.CR], (25 Mar 2021)
@misc{dimitrov2021cuda,
title={CUDA Tutorial — Cryptanalysis of Classical Ciphers Using Modern GPUs and CUDA},
author={Miroslav Dimitrov and Bernhard Esslinger},
year={2021},
eprint={2103.13937},
archivePrefix={arXiv},
primaryClass={cs.CR}
}
CUDA (formerly an abbreviation of Compute Unified Device Architecture) is a parallel computing platform and API model created by Nvidia allowing software developers to use a CUDA-enabled graphics processing unit (GPU) for general purpose processing. Throughout this tutorial, we introduce the CUDA concepts in an easy-to-grasp interactive way. Starting from scratch, we implement a complete stand-alone GPU tool for automatically decrypting ciphertexts (ciphertext-only attack) encrypted by monoalphabetic substitution and columnar transposition ciphers. Throughout this process, we will learn how to architect the tool, what optimizations could significantly empower our routines, why the choice of an adequate metaheuristic is critical, and how to draw sketches to enlighten the design process, by proactively solving upcoming issues.
March 28, 2021 by hgpu
