TRAILS: Efficient data flow tracking through hw-assisted parallelization
Abstract
The goal of this effort is to make currently opaque computing systems transparent by providing high-fidelity visibility into component interactions during system operation at the hardware architecture layer, while imposing minimal performance overhead. The scope of this effort is to develop a novel architecture for efficiently performing dynamic data-flow tracking through hardware-assisted parallelization. The goal includes the creation of an execution environment that provides low-overhead, fine-grained, data-flow tracking (DFT) as a utility for security, privacy, and other applications. Trails aims to exploit the parallelism available in modern architectures to improve DFT performance without the demand for more resources than the ones already required to apply it in line with the application.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 2019
- Accession Number
- AD1081339
Entities
People
- Georgios Portokalidis
Organizations
- Stevens Institute of Technology