SAMPA: Security Analysis and Monitoring to Prevent Abuse of High Performance Computing Environments
- Intelligent Automation
- University of South Florida
- George Mason University
HPC systems should not be abused to carry out cybercrimes or execute improper jobs. Existing solutions such as fingerprinting require runtime data collection and processing, which impacts the performance of HPC. Moreover, existing solutions are ineffective against adversarial scenarios where malicious users intend to hide their behavior patterns. Thus, it is essential to build a powerful offline software analysis tool that can scan/analyze obfuscated software programs to detect illicit behaviors. In Phase II, we developed a prototype of deep-learning-based program analysis toolkit dubbed SAMPA, and evaluated its performance over the publicly available datasets and demonstrated its feasibility. It can provide abuse/misuse prevention and malware detection for HPCs and other similar systems to better assist end users in program security analysis.
- Research Organization:
- Intelligent Automation
- Sponsoring Organization:
- USDOE Office of Science (SC)
- DOE Contract Number:
- SC0018476
- OSTI ID:
- 1845457
- Type / Phase:
- SBIR (Phase II)
- Report Number(s):
- DOE-IAI-18476
- Country of Publication:
- United States
- Language:
- English
Similar Records
Malware detection and analysis
Spy the Lie: Detecting Malicious Insiders