Machine Learning-Based Temperature Prediction for Runtime Thermal Management Across System Components

• Published in: IEEE transactions on parallel and distributed systems Vol. 29; no. 2; pp. 405 - 419
• Main Authors: Kaicheng Zhang, Guliani, Akhil, Ogrenci-Memik, Seda, Memik, Gokhan, Yoshii, Kazutomo, Sankaran, Rajesh, Beckman, Pete
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.02.2018; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Analytical models; Artificial intelligence; Comparative studies; Computational modeling; Cooling; Gaussian process; Gaussian processes; high performance computing systems; High temperature; Lightweight; Machine learning; many-core processors; Migration; Neural networks; Nodes; operating systems; Performance degradation; Performance enhancement; Placement; Predictive models; Regression analysis; Regression models; Thermal management; Thermal modeling; Thermal response; Throttling; Time management; Weight reduction

Elevated temperatures limit the peak performance of systems because of frequent interventions by thermal throttling. Non-uniform thermal states across system nodes also cause performance variation within seemingly equivalent nodes leading to significant degradation of overall performance. In this pa...