A holistic and optimal approach for data center cooling management

Abstract

Abstract-Efficient and reliable operation of today's data centers, which host IT equipment with ever-increasing power density, relies heavily on the cooling system to meet the thermal management needs of the IT equipment with minimal environmental footprint. The dynamic IT workload, together with the spatial variance of cooling efficiencies, creates both temporal and spatial non-uniformities within the data centers. Most data centers use zonal cooling actuators, such as computer room air conditioners (CRAC), to alleviate the local "hot spots". Without proper localized cooling actuation mechanisms, the cooling capacity is usually over-provisioned that leads to waste of energy. To address this problem, we introduce adaptive vent tiles (AVT) for local cooling adjustment, and develop a holistic multivariable model based on the mass and energy balance principles to capture the effects of both zonal and local cooling actuation on the inlet temperatures of the racks that host the IT equipment. A model predictive controller is then proposed to minimize the total cooling power while meeting the thermal requirements of the racks. The zonal and local cooling actuation is coordinated in such a unified framework for the provisioning, transport and distribution of the cooling resources in the data centers. The proposed holistic cooling approach is validated in a production data center. Experimental results indicate that up to 36% of CRAC units blower power can be saved, compared with the state of the art control solution.

Performance evaluation and modeling of active tile in raised-floor data centers: An empirical study on the single tile case

“…Besides the supplying cold air volume, the air flow direction could also be turned by the damper vane to reduce the bypass. Model Predictive Controller (MPC) and Multi-Input-Multi-Output (MIMO) controller were applied to dynamically control AVTs to make the cold air supply adapt to the variable thermal load (Zhou et al, 2011, Zhou et al, 2012. In these systems, the rack inlet temperature was continuously captured and fed back to the controller.…”

Energy efficiency improvement method for data centers based on hybrid proximal policy optimization algorithm

“…Particle swarm optimization was employed to optimize cooling energy consumption. In [14], a model predictive controller (MPC) was constructed to lower cooling system energy consumption by adjusting cooling parameters such as air supply temperature while satisfying the thermal constraints of the data center. In [15], a novel deep reinforcement learning (DRL) framework was proposed to optimize cooling control of data centers.…”

“…The cooling system in current data centers mainly uses two cooling methods: air-cooling and water-cooling. Air-cooling is the primary cooling method for a large number of existing data centers [14,17,25]. Its advantage lies in its low deployment cost and ease of implementation.…”

“…In many data centers, the cooling capacity is typically oversupplied for safe operation, resulting in energy waste. Corresponding research work has also been carried out in the cooling system [13][14][15]. Existing data center construction or renovation overly relies on the transformation of high-efficiency hardware facilities and the selection of natural cooling sources, which are limited and not suitable for all existing or under-construction data centers.…”

An optimization strategy for enhancing energy consumption performance in digital currency miner's building

“…Since the heat transfer of air is low, the flow rate shall be high, which entails an increase in the number of fans. If appropriate control systems are utilized, the energy consumption of fans might be reduced [19]. Since a continuous airflow rate is required between inside and outside, the outdoor climate certainly affects the cooling system's effectiveness [20].…”