Job Scheduling with Machine Speeds for Password Cracking Using Hashtopolis

Abstract

Abstract Due to the current challenges in computer forensics and password cracking, a single GPU is no longer sufficient. Thus, distributed password cracking platforms with dozens of GPUs become a necessity in the race against criminals. In this paper, we show a multi-GPU cracking platform build on Hashcat-based open-source distributed tool Hashtopolis for use in password cracking and computer forensics. We present a mathematical model of the problem, formulating it as a specific case of the problem of scheduling independent jobs on parallel machines with machines speed and makespan criterion. We propose two metaheuristic algorithms based on the Simulated Annealing and Genetic Algorithm method to solve this problem. We employ the algorithms in a computer experiment using real-life password cracking instances and hash functions. The results indicate moderate (4–8%) to considerable (14–38%) improvement in makespan compared to default schedule for most instances. We also show that hash function type affects the improvement. SA and GA show little difference in quality, with SA being slightly better.