Data centers (DCs) are complex organizational and technical infrastructures that assure the performance and reliability of modern information and communication systems. The high installation and operations costs of DCs and the stringent requirements regarding reliability and safety require close attention to the location of this type of facility. This paper proposes a multicriteria decision analysis (MCDA) approach for identifying the most interesting locations to install sustainable DCs, taking into account technical, social, economic, and environmental dimensions. For each of these main dimensions, the evaluation was formulated as a multicriteria sorting problem. These problems were analyzed using the outranking MCDA method ELECTRE TRI through the IRIS software, allowing for uncertainty about the criteria weights. The results are summarized in a graphical form, without attempting to reduce such incommensurable dimensions to a single value.

Data Centers (DCs) are high energy intensive facilities due to the large power requirements of the computing equipment and the infrastructure needed to support it, in particular cooling systems. Nowadays, the capacity and consequently the energy demand of DCs is growing rapidly to meet the digitalization of the society. In order to meet sustainability concerns, the construction of a new DC must address the economic and environmental impacts of the different design options and one of the most crucial choices to be made concerns the location of the DC. This paper presents a brief overview of the current landscape of Portuguese DCs, based on the results of a survey. It also proposes a methodology to identify the best areas to locate a new DC in Portugal, using a new structure for evaluation criteria, based on inputs from industry experts. The relevant data was collected in a Geographical Information System (GIS) and successive filters were applied to progressively reduce the number of candidate sites. Finally, an analysis to the benefits of having DCs on those sites in comparison with their current locations was made, and the results show that the chosen sites to locate these facilities were not the best solution to achieve a more sustainable DC.

A Data Centre is a physical space that groups together computer equipment such as servers, storage arrays, among other equipment. It can be used for storage, processing and data protection. The majority of data centres operate 24 hours a day, and must provide the user with guarantees in terms of security and performance. Energy consumption is therefore permanent and the level of assistance and maintenance high. Data Centres have an ecological impact that is almost invisible to many users. In this paper a computational fluid dynamics (CFD) model of the airflow and thermal distribution of a real data centre room with 208 racks is proposed. Two case studies are presented and simulated with a high thermal load subjected to the minimum and maximum air flow velocity, respectively. The objective is to assess if the computer room air conditioning unit (CRAC) can cool the racks of the data centre in order to efficiently refrigerate all the hot spots.

In recent years, reducing energy consumption has been relentlessly pursued by researchers and policy makers with the purpose of achieving a more sustainable future. The demand for data storage in data centers has been steadily increasing, leading to an increase in size and therefore to consume more energy. Consequently, the reduction of the energy consumption of data center rooms is required and it is with this perspective that this paper is proposed. By using Computational Fluid Dynamics (CFD), it is possible to model a three-dimensional model of the heat transfer and air flow in data centers, which allows forecasting the air speed and temperature range under diverse conditions of operation. In this paper, a CFD study of the thermal performance and airflow in a real data center processing room with 208 racks under different thermal loads and airflow velocities is proposed. The physical-mathematical model relies on the equations of mass, momentum and energy conservation. The fluid in this study is air and it is modeled as an ideal gas with constant properties. The model of the effect of turbulence is made by employing a k-ε standard model. The results indicate that it is possible to reduce the thermal load of the server racks by improving the thermal performance and airflow of the data center room, without affecting the correct operation of the server racks located in the sensible regions of the room. Appl. Sci. 2019, 9, 3850 2 of 30 the world host on their servers organizations with databases, every Internet website, and cloud services. Data centers are consuming more and more electricity worldwide [4][5][6]. In 2012, the entire consumption of data centers was assessed to be near 270 TWh [7]. ICT are responsible for roughly 10% of worldwide power consumption and 13-15% of this consumption is due to data centers [8][9][10][11]. It is expected that this consumption will only grow in the near future, thus challenging sustainability goals [12]. In a study where projections up to 2030 are made, data centers could consume, in the best case scenario 3% and in the worst case scenario 13% of the worldwide electricity production [13]. The main factor of such a high level of energy use is the consumption of air conditioning systems, which, given the type of ICT equipment, could account for 24-60% of the entire data center's energy consumption [14].Data centers are required to follow rigorous technical specifications with the purpose of guaranteeing their optimal operation and security, and engineers need to sure that the ICT equipment should operate without experiencing any loss of performance or interruption. Thus, thermal engineers have a crucial importance in the design of data centers. The reason is that data centers generate a high quantity of heat and need to be refrigerated in order to avert any loss of performance or failures, thus signifying that the power consumption is high [15].Data center rooms are comprised of racks hosting servers with distinct configurations and distinct capacities. To keep the productiv...