Jim Brandt is a researcher in the HPC Development department at Sandia National Laboratories where he is the technical lead of the AppSysFusion project and associated suite of HPC monitoring and analysis tools including the Lightweight Distributed Metric Service (LDMS) HPC monitoring framework. Jim’s research interests are in automating analyses and feedback that can be applied to run time application and system monitoring data to enable more efficient HPC system operation and workflow execution. His focus is currently on the use of statistical and machine learning methods for characterizing the resource needs of applications, the impact of aggregate workloads on shared resource performance, and the impact of degraded resource performance on application and overall workload performance. Jim is on the organizing committee of the Monitoring and Analysis for HPC Systems Plus Applications (HPCMASPA) workshop now going into its 10th year.
HPC Community Collaboration
Machine Learning and Artificial Intelligence
Resource Management and Scheduling