A Data-driven, Multi-setpoint Model Predictive Thermal Control System for Data Centers
- PDF / 1,806,638 Bytes
- 22 Pages / 439.37 x 666.142 pts Page_size
- 7 Downloads / 138 Views
(2021) 29:7
A Data‑driven, Multi‑setpoint Model Predictive Thermal Control System for Data Centers SeyedMorteza Mirhoseininejad1 · Ghada Badawy2 · Douglas G. Down1 Received: 20 June 2020 / Revised: 7 September 2020 / Accepted: 7 October 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract This paper presents a system for jointly managing cooling units and workload assignment in modular data centers. The system aims to minimize power consumption while respecting temperature constraints, all in a thermally heterogeneous environment. Unlike traditional cooling controllers, which may over/under cool certain areas in the data center due to the use of a single setpoint, our framework does not have a single setpoint to satisfy. Instead, using a data-driven thermal model, the proposed system generates an optimal temperature map, the required temperature distribution matrix (RTDM), to be used by the controller, eliminating under/over cooling and improving power efficiency. The RTDM is the resulting temperature distribution when jointly considering workload assignment and cooling control. In addition, we propose the use of model predictive control (MPC) to regulate the operational variables of cooling units in a power-efficient fashion to comply with the RTDM. Within each iteration of the MPC loop, an optimization problem involving the thermal model is solved, and the underlying thermal model is updated. To prove the feasibility of the proposed power efficient system, it has been implemented on an actual modular data center in our facilities. Results from the implementation show the potential for considerable power savings compared to other control methods. Keywords Data center workload assignment · Cooling unit control · Thermal-aware scheduling · Thermal model · Data center power efficiency · Efficient cooling · Model predictive control · Multi setpoint control
* Douglas G. Down [email protected] SeyedMorteza Mirhoseininejad [email protected] Ghada Badawy [email protected] 1
McMaster University, 1280 Main St W, Hamilton, ON, Canada
2
Computing Infrastructure Research Centre, 175 Longwood Rd, Hamilton, ON, Canada
13
Vol.:(0123456789)
7
Page 2 of 22
Journal of Network and Systems Management
(2021) 29:7
1 Introduction Due to opportunities provided by the increasing use of mobile, cloud, and processing services, a large portion of IT investment is toward data center development and expansion [1]. This increase stems mainly from binding mobile applications and services to daily life, and migrating processes from end-user devices (such as laptops or cell phones for the sake of battery life, security, and integrity) to the server-side [2]. These changes have made data centers among the most power-hungry infrastructures [3]. Powering computing devices and cooling them are the two main sources of power consumption in data centers. Reducing power consumption has been a focus of research for a number of years, having been studied at different levels from small electronic componen
Data Loading...
