Dynamic prediction & estimation of intentional failures in HPCs
A. D. Clark, L. M. Tellez, S. Besse, J. M. Absher · 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM) · 2016
High performance computing systems are becoming the norm for daily use and their applications are being known within academia, industry, and government sectors. However, the resilience of these systems is in question for their complex internal structure makes them difficult to trouble shoot - making them vulnerable to intentional failures. Our work addresses this topic by employing dynamic prediction and estimation strategies of observed failures for individual nodal components within the exascale class network. Experiments using a simulated HPC environment convey the efficacy of our process where the results can be directly applied for improved detection and monitoring of cyber anomalies.