Presentation
SIGN IN TO VIEW THIS PRESENTATION Sign In
DPS: Adaptive Power Management for Overprovisioned Systems
SessionResource Management
DescriptionMaximizing performance under a power budget is essential for HPC systems and has inspired the development of many power management frameworks. These can be broadly characterized into two groups: model-based and stateless. Model-based frameworks achieve good performance under a power budget but are highly dependent on the quality of the model and the data used to train it. Stateless frameworks are more robust and require no training, but are generally lower performance. In this paper, we propose a new framework that does not require a model, but does track state in the form of recent power dynamics. We implement this idea and test it on a public cloud running both Spark and HPC jobs. We find when total power demand is low, our framework achieves equivalent performance to prior work, but when power demand is high it achieves a mean 8% performance improvement (with no reliance on a learned model).
Event Type
Paper
TimeTuesday, 14 November 20232:30pm - 3pm MST
Location401-402
Architecture and Networks
Performance Measurement, Modeling, and Tools
Resource Management
TP
