Machine Learning Applications for Data Center Optimization

By J. Gao. 2014

262 citations recorded by the source corpus.

Subfields: Reinforcement Learning, ML Systems & Hardware

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Reinforcement Learning

In ML Systems & Hardware

Citations in the corpus

Quantifying the Carbon Emissions of Machine Learning (2019, methodology): “For example, Google Cloud Services has an average PUE of 1.1, meaning that only 11% of their total energy usage is not used for the servers themselves, a ratio that they have been steadily reducing using Reinforcement Learning [16, 17].”
Data center cooling using model-predictive control (2018, methodology): “Recently, DeepMind demonstrated that it is possible to improve DC power usage efficiency (PUE) using a machine learning approach [13].”
Data Center Energy Consumption Modeling: A Survey (2016, unknown): “12 in 2013 [298].”
The environmental footprint of data centers in the United States (2021, background): “07 has been reported by Google for some of their data centers [55].”
A Tour of Reinforcement Learning: The View from Continuous Control (2018, background): “ly harder instance of LQR and show how it highlights issues of robustness and safety. Consider an idealized instance of \data center cooling," a popularized application of reinforcement learning [32]. Dene the model to have three heat sources coupled to their own cooling devices. Each component”
Demand Response for Home Energy Management Using Reinforcement Learning and Artificial Neural Network (2019, methodology): “A practical pioneer of this approach is Google DeepMind who successfully applied a RL-based scheme to reduce the electricity bill associated with data center cooling by 40% [12]; which is another key factor for motivating the application of RL technology to energy systems.”
Tackling Climate Change with Machine Learning (2019, background): “inforcement learning to optimize cooling centers for Google’s internal servers by predicting and optimizing the power usage effectiveness (PUE), thus lowering HFC emissions and reducing cooling costs [361, 394]. Deep neural networks could also be used for adaptive control in 28 a variety of industri”
Green Algorithms: Quantifying the Carbon Footprint of Computation (2020, background): “[60,86] Carbon Intensity of Energy Production : For a given country and energy mix, the carbon footprint in CO 2 e represents the amount of CO 2 with the same global warming impact as the GHGs emitted, which simpliﬁes the comparison between diﬀerent electricity production methods.”