A Cross-Stack Approach Towards Defending Against Cryptojacking
Nada Lachtar, Abdulrahman Abu Elkhail, Anys Bacha, Hafiz Malik
Abstract
Cryptocurrenices are revolutionizing the way we conduct every day business. Unfortunately, cybercriminals have harnessed this technology for making profit through cryptojacking, the act of maliciously appropriating computational resources for mining cryptocurrencies. In this letter, we explore a general solution for detecting cryptojacking attacks irrespective of the application type. We propose an end-to-end detection solution that leverages lightweight microarchitectural changes designed to track instructions that are commonly used in hash algorithms. An evaluation of our implementation shows negligible performance overhead while testing across a mix of workloads from the SPEC 2006 benchmarks.