Search for: All records

The role of data centers in modern life has expanded rapidly over the past decades. In addition, this expansion has resulted in a significant increase in the share of data centers in total energy consumption of the world. Thus, reliability and energy efficiency have become a common concern in data centers. Information technology equipment (ITE) and cooling infrastructure are the largest power consumer in data centers. The cooling power in a data center depends on the amount of heat dissipated by ITE. Therefore, the thermal design of the ITE impacts not only the ITE power but also affects the infrastructure power and has a significant role in the overall efficiency of data centers. This paper studies the impact of fans location and airflow balancing on the thermal performance and power of a server. A detailed computational fluid dynamic (CFD) model of the server is built, calibrated and validated using experimental test results. Next, impacts of moving fans to the rear side of the chassis on the flow rate and temperature of components are investigated. Special attention is given to controlling airflow through power supplies. 

Optimizing data centers for energy efficiency plays a key role in the successful operation of modern data centers. One important factor is the proper management of airflow. In most air-cooled data centers, the required airflow for cooling of IT equipment is supplied from a raised floor to server racks through perforated tiles. In recent years, different approaches have been implemented to increase the efficiency of air delivery through tiles such as the use of directional tiles, adding understructure scoops or using air dampers [1]. Because the IT load of each rack in the data center is constantly changing due to the processing demands of the IT hardware at a given time, simultaneous manual tuning of the airflow at the panel level is impossible or at least impractical. The amount of airflow delivered to the Cold Aisle Containment (CAC) can be adjusted using Variable Airflow Panels (Dampers) that can be controlled remotely. In this study, we design and optimize a fuzzy control system to control the open area ratio of air dampers in order to adjust the local airflow rate in the ES2 data center. 

The dynamic nature of today’s data centers requires active monitoring and holistic management of all aspects of the facility, from the applications to the air conditioning. The most significant aspect of implementing a dynamic data center is the requirement to actively monitor and manage the infrastructure assets. It is vital to ensure information technology (IT) equipment has access to sufficient air (provisioned) at a proper temperature to assure their optimal and continues operation. Hot air recirculation, elevated fan speed, and hot spots are known consequences of an under-provisioned cold aisle. On the other hand, over-provisioning a cold aisle can lead to a significant loss in energy due to bypass of cooling air and leakages. Besides, the number of active servers in an aisle may be varied by load balancers due to short or long-term IT load changes. This demonstrates the need for an active airflow management scheme that is able to respond to airflow demand in different aisles of a data center. In this study, remotely controllable air dampers are implemented to regulate airflow delivery to a cold aisle containment (CAC) during workload changes in a data center. The energy saving opportunities are investigated and practical considerations are discussed. 

The operation of today’s data centers increasingly relies on environmental data collection and analysis to operate the cooling infrastructure as efficiently as possible and to maintain the reliability of IT equipment. This in turn emphasizes the importance of the quality of the data collected and their relevance to the overall operation of the data center. This study presents an experimentally based analysis and comparison between two different approaches for environmental data collection; one using a discrete sensor network, and another using available data from installed IT equipment through their Intelligent Platform Management Interface (IPMI). The comparison considers the quality and relevance of the data collected and investigates their effect on key performance and operational metrics. The results have shown the large variation of server inlet temperatures provided by the IPMI interface. On the other hand, the discrete sensor measurements showed much more reliable results where the server inlet temperatures had minimal variation inside the cold aisle. These results highlight the potential difficulty in using IPMI inlet temperature data to evaluate the thermal environment inside the contained cold aisle. The study also focuses on how industry common methods for cooling efficiency management and control can be affected by the data collection approach. Results have shown that using preheated IPMI inlet temperature data can lead to unnecessarily lower cooling set points, which in turn minimizes the potential cooling energy savings. It was shown in one case that using discrete sensor data for control provides 20% more energy savings than using IPMI inlet temperature data.