Abstract
This paper addresses the problem of data centers’ cost efficiency considering the potential of reusing the generated heat in district heating networks. We started by analyzing the requirements and heat reuse potential of a high performance computing data center and then we had defined a heat reuse model which simulates the thermodynamic processes from the server room. This allows estimating by means of Computational Fluid Dynamics simulations the temperature of the hot air recovered by the heat pumps from the server room allowing them to operate more efficiently. To address the time and space complexity at run-time we have defined a Multi-Layer Perceptron neural network infrastructure to predict the hot air temperature distribution in the server room from the training data generated by means of simulations. For testing purposes, we have modeled a virtual server room having a volume of 48 m3 and two typical 42U racks. The results show that using our model the heat distribution in the server room can be predicted with an error less than 1 °C allowing data centers to accurately estimate in advance the amount of waste heat to be reused and the efficiency of heat pump operation.
Highlights
Nowadays data centers (DCs) are subjected to significant pressure to perform more efficiently from an environmental perspective towards generating carbon-neutral benefits
Definition of a heat reuse model for DCs allowing them to estimate in advanced the amount of generated waste heat and the impact on the efficiency of the heat pump operation; Definition of Computational Fluid Dynamics (CFD) models to simulate the thermodynamic processes inside the server room and estimate the temperature of the hot air generated; Development of neural networks algorithm to predict the heat distribution in the server room from training data generated using the CFD simulations, making our model feasible for near real-time decision making; By using the proposed approach, the DC operators will be able to accurately forecast the temperature of the hot air recovered from the server room and the amount of waste heat that might be reused
Case, the simulations the room flow ofaiming fluids in an area defined by boundary simulations result of numerous configurations of the single simplified setup will constitute the Applied in DC thermal distribution simulation, CFD tools report an error of about 1 degree Celsius training data set real for the neural network willregarding be used atserver run time for compared to the environment, based being prediction suited for process, decisionwhich analysis room decision making
Summary
Nowadays data centers (DCs) are subjected to significant pressure to perform more efficiently from an environmental perspective towards generating carbon-neutral benefits. Definition of a heat reuse model for DCs allowing them to estimate in advanced the amount of generated waste heat and the impact on the efficiency of the heat pump operation; Definition of Computational Fluid Dynamics (CFD) models to simulate the thermodynamic processes inside the server room and estimate the temperature of the hot air generated; Development of neural networks algorithm to predict the heat distribution in the server room from training data generated using the CFD simulations, making our model feasible for near real-time decision making; By using the proposed approach, the DC operators will be able to accurately forecast the temperature of the hot air recovered from the server room and the amount of waste heat that might be reused They will be able to compare and contrast additional investment costs with incremental revenues that can be achieved from valorizing forecasted waste heat.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have