GPU-accelerated brute force attacks on ZIP files
Abstract
Powerful and programmable Graphics Processing Units have become commonplace in consumer-grade computers sold or built today. These GPUs have not only made it possible for users to play the latest video games, but also conduct high performance computing at a relatively inexpensive cost. The CUDA Framework, by NVIDIA, allows users to create or run highly parallel programs that run on a computer's GPU. One application of highly parallel computing is for brute forcing passwords. This research analyzes the performance of NVIDIA GPU's for cracking the passwords of encrypted files stored within ZIP archives. The research conducted aims to fill in gaps in existing research. Specifically, this research examines the relationship that the number of encrypted files has on a brute force time. Additionally, this research compares performance across different encryption types and file sizes. The results of the research were significant, adding to the knowledge of GPU-accelerated ZIP file brute force attacks.
Degree
M.S.
Advisors
Smith, Purdue University.
Subject Area
Information Technology|Computer science
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.