Detecting and Identifying Applications by Job Signatures

Authors: Jie Li (Texas Tech University), Brandon Cook (Lawrence Berkeley National Laboratory (LBNL)), and Yong Chen (Texas Tech University)

Abstract: Knowing the applications of jobs running in high-performance computing (HPC) systems is invaluable for administrators. This research aims to detect and identify applications through job signatures built upon monitoring traces obtained from the LDMS monitoring infrastructure on Cori. By constructing job signatures and applying machine learning models to them, we will be able to detect and identify job applications without user intervention. In addition to application names, job signatures offer the potential to study workloads by computation motif.

Best Poster Finalist (BP): no

Poster: PDF
Poster summary: PDF


Back to Poster Archive Listing