China 
Africa 
Explore chapters and articles related to this topic
Basic IT for Radiographers
Published in Alexander Peck, Clark’s Essential PACS, RIS and Imaging Informatics, 2017
Servers are the devices that run centralised applications, such as the backend functionality of a PACS or RIS, and in most cases are housed within a dedicated environment managed by the local IT department, known as the server room. Servers are effectively powerful customised workstations and can either be dedicated (known as ‘pizza box’ servers owing to their shape) or virtual (running as virtual instances on shared hardware). Dedicated pizza box servers are measured in physical size in ‘U’s, with 1 U being 4.4 cm of height in a standard server rack, with cooling requirements measured in British thermal units (BTUs). An IT department uses the total number of BTUs output by all servers in a given area to calculate the volume of air conditioning required. Within the server room, when adding or updating a dedicated server, considerations include whether there is sufficient power available, sufficient cooling, and whether any infrastructure (network, cabling, physical space) is suitable for the revised requirements. Virtual servers can be measured in many ways, but commonly with processor ‘seats’ being indicated (the more of these seats, the more ‘powerful’ the virtual server can be made). As virtual servers can share physical hardware in the server room (and so have lower power, cabling, and cooling requirements), they are commonly preferred by IT departments; however, as PACS and imaging informatics applications have heavy demands on hardware, debate is ongoing as to whether this is the best option, or whether dedicated pizza box servers remain the better choice in the long term, despite the higher physical space, power, and overall cooling requirements. Pizza box servers with associated racks of hard disk storage space are shown in Fig. 2.1: a typical 2010’s era installation.
Airflow uniformity and energy consumption investigation on data centers enhanced by jet fans
Published in Science and Technology for the Built Environment, 2023
Xin Wang, Yingrui Zhu, Baolian Niu
This section will compare and analyze the measured results with the simulation results to verify the rationality and accuracy of the model, boundary condition setting, numerical solution and mesh division. The experimental DC is located in Nanjing, Jiangsu Province, China. Figure 5 shows the arrangement in the experimental DC. The layout and dimensions of the model constructed in this paper are consistent with the experimental DC. The CRACs are 2 sets of Hiross S52 series precision air conditioners for the server room. Six T-type thermocouples as the temperature sensor were set at the height positions of 0.4 m, 1.2 m and 1.6 m from the floor at the front and rear doors of each rack and connected to Agilent data acquisition loggers. The layout of the measuring points is shown in Figure 6. Temperatures of the rack inlet and outlet are measured simultaneously.
Green engineering of Hong Kong and China's first data centre certified to LEED-CS 2009 Platinum
Published in HKIE Transactions, 2018
Helen Cochrane, Fredrick Leong, Steven So, George Or, Melvyn Lai
FDC2 is designed to comply with the Uptime Institute’s Tier III and IV requirements, which is the highest rating measuring the quality and reliability designed in a data centre. It will support non-stop business throughout the year for end-users, with no downtime of the systems during fault or maintenance. As a stratified type thermal energy storage system, this innovative “cooling battery” (first of its kind in Hong Kong) is designed to maintain the server room at a stable temperature and humidity within the operating range of ASHRAE during utility power interruptions (Figure 7). The cooling battery provides a total of 3,600,000 litres of chilled water for 42 minutes of back-up time to override up to six cycles of chiller restart. This is not just an essential feature to help FDC2 achieve the continuous cooling requirements of the Uptime Institute’s Tier III and IV design parameter, but also to allow FDC2 to achieve a lower power usage effectiveness (PUE) (the ratio of total energy entering the data centre to energy used by IT equipment inside the data centre) by leveraging the potential of server room temperature elevation. As water density is inversely proportional to temperature, the cold water can be completely separated from the hotter, lighter water at the top with a thermocline. The designed cold water temperature and volume can then be secured during the operation at all times and scenarios to feed the data centre cooling system, while the return hot water is trapped in the upper layer.
A tool for generation of stochastic occupant-based internal loads using a functional data analysis approach to re-define ‘activity’
Published in Journal of Building Performance Simulation, 2021
The study has highlighted a number of practical issues that are worth considering in more depth. The first is normalization of the data. For this study, the nominal floor areas per meter have been used in order to be consistent with the input potentially required for BES. Normalization of energy demand by floor area, whilst standard, is not always robust and must be applied with caution. For example, the electrical load of a server room depends on how the servers are stacked and can vary significantly even for similar floor areas. Equally, the use of floor area for an office canteen can yield very different normalized energy demand dependent on whether the kitchen is considered on its own or the eating areas are included.