Neural Network Modeling in Model-Based Control of a Data Center

This paper presents the development of a neural network model of the server temperature to be used in model-based control of a data center. Data centers provide the optimal environments for operation of servers and storage devices. Conventionally, computational fluid dynamics (CFD) has been used to model the dynamic and complex environment of the data center. However, the drawback of this approach is its computational inefficiency. The effects of changing a single input may take an entire day to compute. Thus the CFD model is not well suited for model-based control. Instead, we propose to use an artificial Neural Network (NN) model which predicts server temperatures in significantly less time. In addition, this NN model has the capability of learning the environment in the data center by adapting its parameters in real time based on sensor data continuously taken from the data center. This work discusses the current development of the neural network, work being done at the University of Texas at Arlington, to include modeling of transient conditions, or time related changes, using data generated in a test bed Data Center at SUNY Binghamton.