Presentation
Preparation and Optimization of a Diverse Workload for a Large-Scale Heterogeneous System
SessionState of the Practice
Event Type
Paper
TP
Data Analytics
MPI
Performance
Resiliency
Resource Management
State of the Practice
TimeTuesday, 19 November 20194:30pm - 5pm
Location205-207
DescriptionProductivity from day one on supercomputers that leverage new technologies requires significant preparation. The institution procuring a novel system architecture often does not have enough people with all the requisite knowledge and skills to prepare for it. Thus, institutions have recently employed the "Center of Excellence" (CoE) concept to prepare for systems such as Summit and Sierra, currently the top two systems in the Top 500.
This paper documents CoE experiences preparing a workload of diverse applications and math libraries for a heterogeneous system. We describe our approach to this preparation, including our management and execution strategies, and detail our experiences with and reasons for employing different programming approaches. Our early science and performance results show the project enabled significant early seismic science with up to a 14x throughput increase over Cori. In addition to our successes, we discuss our challenges and failures so others may benefit from our experience.
This paper documents CoE experiences preparing a workload of diverse applications and math libraries for a heterogeneous system. We describe our approach to this preparation, including our management and execution strategies, and detail our experiences with and reasons for employing different programming approaches. Our early science and performance results show the project enabled significant early seismic science with up to a 14x throughput increase over Cori. In addition to our successes, we discuss our challenges and failures so others may benefit from our experience.
Download PDF
Archive
Authors
