Real-Time Control for Power Cost Efficient Deep Learning Processing With Renewable Generation

• Published in: IEEE access Vol. 7; pp. 114909 - 114922
• Main Authors: Kang, Dong-Ki, Youn, Chan-Hyun
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Artificial intelligence; Artificial neural networks; Back propagation networks; Clusters; Deep learning; Electricity; Electricity consumption; Electricity pricing; frequency control; Graphics processing units; Machine learning; mathematical programming; Micromechanical devices; nonlinear control systems; Operating costs; Performance degradation; Power consumption; Power control; Power demand; power engineering computing; Power management; Predictive control; Process control; Quadratic programming; Real time; Real-time systems; renewable energy sources; Training

The explosive increase in deep learning (DL) deployment has led GPU power usage to become a major factor in operational cost of modern HPC clusters. The complex mixture of DL processing, fluctuated renewable generation, and dynamic electricity price impedes the elaborate GPU power control, so as to...