Flexible hyperscale data centre

Abstract

A computer implemented method for designing a flexible hyperscale data centre comprises designing a configurable structural framework model for the data centre 402 with a defined building footprint an

Description

Field of the Invention The field of the invention relates to hyperscale data centres. The invention is applicable to, but not limited to data centres with increased area for data halls. Background of the invention A hyperscale data centre is a large-scale facility typically used by major technology companies with high cloud computing demand, Al, and large data processing requirements. Ideally, the end user (customer) of a data centre would be certain from the outset. However, often data centres are designed speculatively where the developer will target a range of the hyperscale clients to sell the computing space to. This often results in uncertainty on end rack layouts which will differ from design stage delivery. As illustrated in Figures 1(a)-(c) most data halls, 100, use hot aisle, 104, containment. Figure 1 (a) shows a 6MWdata hall, 1(b) a 12MWdata hall and 1(c) an 18MWdata hall. The key for figure 1 is as follows: 100 - Overall data hall 101 - CRAG, CRAH or FWU 102-Cold Aisle 103 - Cooling Corridor 104-Hot Aisle 105 - Services Riser 110 - Structural Column 130 - Server Racks Each data hall is provided with a plurality of server racks, 130, with Computer room air handler (CRAH), Fan wall units (FWU) or Computer room Air conditioning units CRAC, 101, positioned around the boundary of the data hall in a separate cooling corridor. Hot aisle containment is a data centre cooling strategy that optimizes airflow by arranging server racks 130, in alternating hot-104 and cold-102 aisles, then physically enclosing the hot aisle to trap the heat exhausted from the rear of the servers-130 / ; this contained hot air is then directed back to the cooling system via a false ceiling. While the cold aisle-102 provides a consistent supply of cool air to the front of the racks, thereby preventing hot and cold air mixing, significantly improving cooling efficiency, reducing energy consumption, and ensuring more stable operating temperatures for IT equipment. Figure 2 illustrates the typical data centre floor plans and roof plans, and arrangement that must be coordinated as part of the flexible hyperscale data centre design process. The typical floor layout illustrates the following; 201 shows the admin / office areas which contain operations control rooms and general workplace rooms, 202 shows stairwells, 203 are all generators which inform the electrical lineups, 204 are electrical rooms which may be switchgear cubicles, UPS rooms or IDF rooms, 205 shows storage rooms which vary dependant on the hyperscale end-user. The typical roof layout illustrates the following; 206 are rooftop electrical communication rooms, 207 shows air-cooled chillers which selections depend on the temperature of the data hall, 208 are mechanical plant rooms which tend to be the same size throughout different end-users and contain equipment which maintains the function of the CHW systems such as filtration equipment, chemical dosing and pressurisation, 209 shows airhandling units which provide fresh air ventilation and have slight variance between end-users, 210 shows buffer vessels which size will vary depending on the end-user. Currently, typical data centre designs are highly specific and not adaptable for various end-user applications. There is a lack of flexibility to accommodate a range of end-user designs efficiently within a single building footprint. There is also limited information available to utilise or build upon to readily deduce achieving a data centre design in the speculative build market, capable of catering to diverse hyperscale data centre requirements. Accordingly, there is a need for a hyperscale flexible data centre. This proposed invention addressed this by designing the data centre to ensure seamless adaptability to diverse client requirements and to increase the overall data hall area in the data centre. Summary of the invention According to an embodiment there is provided a computer implemented method for designing a flexible hyperscale data centre, where the data centre comprises a plurality of data hall configurations and arrangements of server racks in each data hall with increased data hall provision comprising the steps of: designing a configurable structural framework model for the data centre with a defined building footprint and a data centre roof, wherein the structural framework model comprises variable parameters related to: access requirements within each data hall, hyperscale general storage area requirements within each data hall, data hall height, data hall width, data hall depth, data hall access, floor loading capacity within the data halls, and in parts of the data centre external to the data halls, hot and cold aisle containment requirements for each data hall, air circulation pathways within each of the data halls and fire ratings for each data hall; determining a plurality of alternative configurations for the arrangement of: mechanical components and electrical components providing cooling and ventilation to each of the plurality of data halls, electrical components providing power to the server racks in each of the data halls, and plumbing services and components in each of the plurality of data halls, in accordance with the parameters of the configurable structural framework model for each of the plurality of data halls; setting a power density for each of the plurality of data halls, determined by the arrangement of electrical components in each of the data halls; modelling a plurality of different air supply temperature in the range 27-35°C, a plurality of pressure values in the range 0 to 15Pa and a plurality of relative humidity values in the range of 5-80%, for each of the various mechanical, electrical and plumbing configurations in the plurality of electrical rooms, data halls, across a range of different operating conditions for each of the alternative configurations, and determining a Power Usage Effectiveness for each operating conditions across the range of alternative configurations and temperature, pressure and humidity values; using the results from the modelling step to create an output of multiple different data centre configurations within the building footprint, wherein each of the output data centre configurations will define the configuration of the mechanical services and components, the electrical services and components, and the plumbing services and components, in each of the plurality of data halls and any other rooms associated within the data centre facility, so that the data hall space within the building footprint is at least 45%. Preferably, the power usage effectiveness value is minimised to be less than 1.2 for each data hall configuration. Further preferably, the power usage effectiveness is minimised by optimising the components related to providing cooling, and then optimising the cooling temperatures within the cooling components. In a preferred embodiment, the components for cooling are selected to maximise the use of free cooling in each data hall. Preferably, the value of the power usage effectiveness is reduced to under 1.2 by adjusting a chilled water flow and return temperatures for rack systems in the data centre. Further preferably, wherein the mechanical components comprise at least one of CHW, AHU, CRAC, CRAH and FWU units. In a preferred embodiment the electrical components comprise one or more of power supply units, backup generators, uninterruptable power supply units. Preferably, the plumbing services and components comprises fire suppression system and water based cooling systems. Further preferably, the electrical components are configured to provide a redundancy of components in each data hall. Preferably, the redundancy for each data hall is N+1, 2N, or 2(N+1), where N is the number of components with no redundancy. Preferably the components for cooling of the data centre are selected in accordance with local climate information. Further preferably, the configurable structural model of the data centre further comprises parameters representing one or more chillers to be positioned on the roof of the data centre. In an embodiment, there is also provided a flexible hyperscale data centre designed using the method described above. These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter. Brief description of the drawings Further details, aspects and embodiments of the invention will be described, by way of example only, with reference to the drawings. In the drawings, like reference numbers are used to identify like or functionally similar elements. Elements in the figures are illustrated for simplicity and clarity and have not necessarily been drawn to scale. Figure 1(a)illustrates a typical Data Hall Arrangement at IT loadings of 6MW: Figure 1 (b)illustrates a typical Data Hall Arrangement at IT loadings of 12MW: Figure 1(c): illustrates a typical Data Hall Arrangement at IT loadings of 18MW: Figure 2(a): illustrates a typical floor layout for a Data Centre; Figure 2(b) illustrates a typical roof layout for a data centre: Figure 3: illustrates a flow Chart of Flexible Data Centre Design Stages according to an embodiment;. Figure 4 illustrates an alternative flow chart for data centre design according to an embodiment; Figure 5: Illustrates a 12MW Data Hall Rack &Floorplan Configuration for scenario 1;. Figure 6: illustrates a 12.75MW Data Hall Rack Configuration for scenario 2. Detailed Description List of tables Table 1: Key Data Centre Features in Order of Importance to Flexible Design. Table 2: CSA discipline variables. Table 3: Electrical discipline variables. Table 4: Facility discipline variables. Table 5: Fire discipline variables. Table 6: Mechanical discipline variables. Table 7: Security access and other disciplines variables. Table 8: Flexible Data Centre Combinations. As speculative data centre designs often lack knowledge of the end customer at the outset, an embodiment was directed to creating flexible and adaptable design solution considerations, with improved building layout and maximising the useable area of the data centre for data halls, as well with minimal distribution, cost, and program impact. Extensive research was undertaken to understand the main data centre hyperscale customer requirements from a Mechanical, Electrical, and Plumbing (MEP) design perspective. Each discipline focuses on the following. Mechanical Systems: • This primarily focuses on cooling and ventilation in each of the data halls, as well as the other areas of the data centre. Data centres generate a lot of heat, so robust CHW &HVAC (heating ventilation and air conditioning) systems are needed to maintain optimal temperatures and humidity levels throughout the data centre. This includes CRAG, CRAH or FWU units, as well as airflow management strategies like hot and cold aisle containment. Electrical Systems: • This includes the primary power supply, backup generators, uninterruptible power supplies (UPS), and power distribution units (PDUs). Redundancy is key, ensuring continuous power provision to the data halls even during power outages. • Grounding and surge protection are also essential to protect sensitive equipment. Plumbing Systems: • While less prominent, plumbing is still important for fire suppression systems (sprinklers or clean agent systems), humidification and for cooling systems that use water. A detailed design matrix was created and all the main hyperscale customer design requirements were analysed in detail to establish if the design solutions could be applied across a multitude of different data centre projects. 1. Shell &Core / Data Hall Items 2. Highly Customer Specific 3. Subject to Minor Customer Alignment Rack density LV Output Chillers Racks per row STS (Static Transfer Switches) Cooling Units Structural Allowance UPS MUPS PDU Air Handling Units Data hall temperature Generators MV Switchgear T ransformers LV Input Table 1: Key Data Centre Features in Order of Importance to Flexible Design Table 1 shows the key Data Centre Features in order of importance to flexible design, the first column ’Shell and Core / Data hall items’ defines the design criteria to determine the optimal building shell &core, structure, and data hall size to accommodate a wide range of end clients. These criteria include rack density, racks per row and a structural allowance. The second column, 'Highly Customer Specific,' identifies elements, such as low voltage LV output, static transfer switches (STS), UPS, PDU, and data hall temperature, that are addressed by considering the maximum requirements across all potential clients, creating a flexible baseline. Finally, the third column, 'Subject to Minor Customer Alignment,' covers other items like chillers, cooling units, air handling units, generators, MV switchgear, transformers, and LV input. These elements are typically designed with a degree of standardisation but may require minor adjustments to align with specific client operational preferences, further enhancing the data centre's adaptable nature through the hybrid approach. The technological advance sought was the improvement in knowledge and capabilities toward data centres through the development of a flexible and adaptable data centre design, with improved usage of the data centre, by maximising the space available for use as data hall. For a designer who is configuring the mechanical, electrical and plumbing systems and components, the most efficient design process occurs when the end user of the data centre is identified at the start of the project, before the data centre design is finalised. This allows the layout and infrastructure of the data centre, and each of the data halls to be tailored to exact customer requirements. However, in the case of a flexible data centre design, the end user may not be known at the outset. To address this, a hybrid design approach is implemented, often trying to incorporate select items stated within the end users Basis of Design (BoD) or Statement of Qualification (SoQ) whilst allowing the facility to adapt to different customer needs at various stages of the design process. For example, if one hyperscale customer has a minimum ceiling height in data halls of 3.5m and the other has 4m, the most stringent requirement will be considered so that the design can work for both hyperscale customers. This differs from traditional data centre designs, which are often built with fixed specifications for a known client. A flexible data centre incorporates modular systems, scalable infrastructure, and adaptable space planning, enabling it to accommodate multiple end users with varying operational and design requirements. This project aimed to improve the conventional approach to data centre design by creating a unique template capable of accommodating various hyperscale data centre requirements within a single building footprint, whilst increasing the area in the data centre taken up by the data halls. According to an embodiment there is provided a computer implemented method for designing a flexible hyperscale data centre, where the data centre comprises a plurality of data hall configurations and arrangements of server racks in each data hall with increased data hall provision comprising the steps of: designing a configurable structural framework model for the data centre with a defined building footprint and a data centre roof, wherein the structural framework model comprises variable parameters related to: access requirements within each data hall, hyperscale general storage area requirements within each data hall, data hall height, data hall width, data hall depth, data hall access, floor loading capacity within the data halls, and in parts of the data centre external to the data halls, hot and cold aisle containment requirements for each data hall, air circulation pathways within each of the data halls and fire ratings for each data hall; determining a plurality of alternative configurations for the arrangement of: mechanical components and electrical components providing cooling and ventilation to each of the plurality of data halls, electrical components providing power to the server racks in each of the data halls, and plumbing services and components in each of the plurality of data halls, in accordance with the parameters of the configurable structural framework model for each of the plurality of data halls; setting a power density for each of the plurality of data halls, determined by the arrangement of electrical components in each of the data halls; modelling a plurality of different air supply temperature in the range 27-35 °C a plurality of pressure values in the range 0 to 15Pa and a plurality of relative humidity values in the range of 5-80%, for each of the various mechanical, electrical and plumbing configurations in the plurality of electrical rooms, data halls, across a range of different operating conditions for each of the alternative configurations, and determining a Power Usage Effectiveness for each operating conditions across the range of alternative configurations and temperature, pressure and humidity values; using the results from the modelling step to create an output of multiple different data centre configurations within the building footprint, wherein each of the output data centre configurations will define the configuration of the mechanical services and components, the electrical services and components, and the plumbing services and components, in each of the plurality of data halls and any other rooms associated within Data centre facility, so that the data hall space within the building footprint is at least 45%. Flow charts of the steps taken in an embodiment is shown in Figures 3 and 4. Figure 3 shows a method 300 for the flexible design of a data centre, with maximised data hall space in the data centre. Method 300 has the following steps: at step 302 the range of end users is defined. At step 304 the flexible data hall, and shall and core of the building is defined based on a structural grid, weight density temperature and UPS topology amongst other features. At step 306 the building footprint is determined. At Step 308 the speculation is made on the mechanical electrical and plumbing areas which require minor customer alignment. At step 310 flexible but efficient scenarios are made by altering temperature and UPS topologists. At step 312 a limitation on portfolio wide plant procurement is performed to take account of local building codes, different climates and different customers leases across projects. Finally at step 314 a combined flexible optimal and portfolio speculative design is completed for each end user that can use the same building footprint The first step, 302, is where the potential end-users of the data centre are selected. At step 304, any of the features which cannot be changed later down the design stage are set. At 306 At step 308 speculation is made on MEP areas which require minor customer alignment. In step 310 flexible but efficient scenarios are made by altering temperature and UPS topologies. The cooling capacity redundant configurations to further enhance reliability. Common UPS configurations include N+1 redundancy (an extra UPS module is added to the system), 2N redundancy (two independent UPS systems provide power to the equipment) and 2(N+1) redundancy (combines the concepts of 2N and N+1). Cooling capacity is determined by available white space in the data halls and basing off the maximum W / m2 off worst case client requirements. In 310 - One way to measure energy efficiency in data centres is with an index called Power Usage Effectiveness, PUE, which is the ratio between IT power and Total facility power. PUE _ Total Data Centre Facility Energy Consumption IT Equipment Energy Consumption ' ' Preferably, the power usage effectiveness value is minimised to be less than 1.2 for each data hall. Further preferably, the power usage effectiveness is minimised by optimising the components related to providing cooling, and then optimising the cooling temperatures within the cooling components. Optimising PUE in data centres involves adjusting chilled water flow and return temperatures to match varying rack inlet requirements, thereby enhancing cooling efficiency and allowing more use of free cooling. Preferably, the value of the power usage effectiveness is reduced to under 1.2 by adjusting a chilled water flow and return temperatures for rack systems in the data centre. In data centres, free cooling is the practice of dissipating heat without artificially cooling air or water. Typically, free cooling systems work by collecting air or water from the ambient environment, then circulating it into data centre server rooms or individual server racks. Free cooling is distinct from mechanical cooling, which relies on refrigerants and compressors to cool air or liquid. The main benefit of free cooling systems relative to mechanical cooling is simple: Free cooling uses much less energy. In turn, it can boost data centre power efficiency and sustainability. Free cooling drastically lowers chiller electrical consumption. Without free cooling, chillers typically account for 40-50% of data centre power usage. Therefore, maximizing free cooling directly improves Power Usage Effectiveness (PUE) and generates significant energy cost reductions. In a preferred embodiment, the components for cooling are selected to maximise the use of free cooling in each data hall. Increasing the temperature in chilled water systems reduces chiller and pump energy consumption, leading to overall efficiency gains. Modifying UPS lineups, such as implementing ECO mode or modular systems, improves power efficiency by reducing conversion losses and matching capacity to actual IT loads as UPS systems generally achieve their highest efficiency within a specific load range, improving the UPS efficiency will improve PUE. Together, these strategies lead to lower energy costs, enhanced flexibility, and a more sustainable data centre operation. Other examples of improving efficiency would be to ensure equipment selections for chillers, CRAHs etc are only accepted if they meet efficiency targets which align to the PUE calculation. Figure 4 is an alternative flowchart of the method according to an embodiment. Fig. 4 is a flowchart of an example process 400. In some implementations, one or more process blocks of Fig. 4 may be performed by a device. As shown in Fig. 4, process 400 may include designing a configurable structural framework model for the data centre with a defined building footprint and a data centre roof, where the structural framework model may include variable parameters related to: access requirements within each data hall, hyperscale general storage area requirements within each data hall, data hall height, data hall width, data hall depth, data hall access, floor loading capacity within the data halls, and in parts of the data centre external to the data halls, hot and cold aisle containment requirements for each data hall, air circulation pathways within each of the data halls and fire ratings for each data hall (block 402). For example, device may design a configurable structural framework model for the data centre with a defined building footprint and a data centre roof, where the structural framework model may include variable parameters related to: access requirements within each data hall, hyperscale general storage area requirements within each data hall, data hall height, data hall width, data hall depth, data hall access, floor loading capacity within the data halls, and in parts of the data centre external to the data halls, hot and cold aisle containment requirements for each data hall, air circulation pathways within each of the data halls and fire ratings for each data hall, as described above. As also shown in Fig. 4, process 400 may include determining a plurality of alternative configurations for the arrangement of: mechanical components and electrical components providing cooling and ventilation to each of the plurality of data halls, electrical components providing power to the server racks in each of the data halls, and plumbing services and components in each of the plurality of data halls, in accordance with the parameters of the configurable structural framework model for each of the plurality of data halls; setting a power density for each of the plurality of data halls, determined by the arrangement of electrical components in each of the data halls (block 404). For example, device may determine a plurality of alternative configurations for the arrangement of: mechanical components and electrical components providing cooling and ventilation to each of the plurality of data halls, electrical components providing power to the server racks in each of the data halls, and plumbing services and components in each of the plurality of data halls, in accordance with the parameters of the configurable structural framework model for each of the plurality of data halls; setting a power density for each of the plurality of data halls, determined by the arrangement of electrical components in each of the data halls, as described above. As further shown in Fig. 4, process 400 may include modelling a plurality of different air supply temperature in the range 27-35°C a plurality of room pressure values in the range 0 to 15Pa and a plurality of relative humidity values in the range of 5-80%, for each of the various mechanical, electrical and plumbing configurations in the plurality of electrical rooms, data halls, across a range of different operating conditions for each of the alternative configurations, and determining a power usage effectiveness for each operating conditions across the range of alternative configurations and temperature, pressure and humidity values (block 406). For example, device may model a plurality of different air supply temperature in the range 27-35°C a plurality of room pressure values in the range 0 to 15pa and a plurality of relative humidity values in the range of 5-80%, for each of the various mechanical, electrical and plumbing configurations in the plurality of electrical rooms, data halls, across a range of different operating conditions for each of the alternative configurations, and determining a power usage effectiveness for each operating conditions across the range of alternative configurations and temperature, pressure and humidity values, as described above. As also shown in Fig. 4, process 400 may include using the results from the modelling step to create an output of multiple different data centre configurations within the building footprint, where each of the output data centre configurations will define the configuration of the mechanical services and components, the electrical services and components, and the plumbing services and components, in each of the plurality of data halls and any other rooms associated within Data centre facility, so that the data hall space within the building footprint is at least 45% (block 408). For example, device may use the results from the modelling step to create an output of multiple different data centre configurations within the building footprint, where each of the output data centre configurations will define the configuration of the mechanical services and components, the electrical services and components, and the plumbing services and components, in each of the plurality of data halls and any other rooms associated within data centre facility, so that the data hall space within the building footprint is at least 45%, as described above. Although Fig. 4 shows example blocks of process 400, in some implementations, process 400 may include additional blocks, fewer blocks, different blocks, or differently arranged blocks than those depicted in Fig. 4. Additionally, or alternatively, two or more of the blocks of process 400 may be performed in parallel. In a preferred embodiment, the mechanical components comprise at least one of CHW, AHU, CRAC, CRAH and FWU units. In a further preferred embodiment, the electrical components comprise one or more of power supply units, backup generators, uninterruptable power supply units.. Preferably, the plumbing services and components comprises fire suppression system and water based cooling systems. Further preferably, the electrical components are configured to provide a redundancy of components in each data hall. In a preferred embodiment, the redundancy for each data hall is N+1, 2N, or 2(N+1), where N is the number of components with no redundancy. Further preferably, components for cooling of the data centre are selected in accordance with local climate information. In an embodiment, the configurable structural model of the data centre further comprises parameters representing one or more chillers to be positioned on the roof of the data centre. In an embodiment there is provided flexible hyperscale data centre designed using the method of any preceding claim. An embodiment of the invention was concerned with how to achieve a solution that provided maximum flexibility to meet a variety of solutions on different projects. Uncertainty was due to the end customer often not being known at the outset of the project of speculative data centre design, requiring the design solutions to be adaptable. There was technological uncertainty surrounding how to achieve a higher IT load density whilst maintaining significant temperatures within the data halls. This is an issue for standard data centres as the maximum temperature the data hall can reach is defined by each end user; it is a contractual element that this temperature must not be exceeded. The increase in density added significant thermal pressure to the data halls, creating uncertainty due to the additional heat load. A flexible technical solution was needed to optimise cooling efficiency and accommodate high server inlet temperatures. This technological uncertainty is compounded by the necessity to identify an optimal solution that can accommodate the varying requirements of the key components in a datacentre which cooling system also need to accommodate. As outlined in the methodology; in order to achieve all of steps 302-314, or steps 402-408 the following design matrix should be taken place to achieve universal design for hyperscale data centre. The present invention proposes a detailed design matrix, as shown in tables 2 to 7 below with numerous design solutions evaluated and assessed ranging from electrical topographies, mechanical cooling solutions and fire engineering solutions. Examples for each of these are as follows. • Electrical topographies - Combined or Separated topologies were considered. In a combined setup, IT and mechanical cooling loads share the same power system. A separated configuration provides independent power systems for mechanical cooling loads. The separated approach was preferred for its greater flexibility in accommodating regional and customer-specific needs. • Mechanical cooling solutions - Required airflow in data halls, types of cooling system, size of buffer tanks and redundancy of air supply can all vary between end users. • Fire engineering solutions - In general there are two types of fire suppression accepted by hyperscale data centre customers. These are sprinklers which require a large storage tank within the building, the other is a gas suppression system which requires bottles of gas to be stored in centralised location or local to each room. Following tables from 1-7, describes clear holistic components to consider for a flexible design in a hyperscale data centre. CSA Data Hall [ )ata Hall CSA Data Hall [ )ata Hall Height CSA Data Hall [ )ata Hall access CSA Data Hall / F Building loor Loadings CSA Data Hall F F tot Aisle Containment Requirement CSA Data Hall ( Zold Aisle CSA Data Hall F r Minimum circulation aisles or naterial movement pathways ■round IT equipment CSA Admin / F Support / Data Hall ire Rating Table 2 Table 2 outlines the Civil, Structural &Architectural (CSA) elements critical to the data hall's adaptability. Recognising that diverse clients possess varying operational needs; a hybrid design approach was implemented. For instance, in determining data hall height, the maximum required dimension across all client specifications was adopted. This ensures the data hall can accommodate the most demanding requirements. Similarly, data hall access was designed to facilitate the installation and removal of all potential client equipment. Floor loadings were calculated to support the heaviest anticipated equipment configurations. Hot and cold aisle containment strategies were developed to allow for flexible implementation based on client preferences. This approach ensures that the fundamental CSA infrastructure is designed with maximum flexibility in mind, enabling efficient customisation for each client. The building itself (walls, ceilings, floor layout) limits where MEP equipment can be placed. The solution must work within these physical constraints. All these factors make finding the optimal building solution complex. Electrical Data Hall Data Hall Electrical Data Hall Power Distribution strategy Electrical Data Hall PDU / PDC Location Electrical Data Hall PDU Corridor Requirement Electrical Data Hall Max. Rack in Row Electrical Data Hall IT Density Electrical Data Hall IT per Rack Electrical Data Hall IT Rack Dimension Electrical Data Hall IT Cabinet Notes Electrical Electrical Electrical Electrical Electrical Utility Supply (HV) Electrical Electrical Power Transformers (HV / MV) Electrical Electrical Primary MV switchgear Electrical Electrical Secondary MV Switchgear Electrical Electrical IT power strings Electrical Electrical Network power strings Electrical Electrical Batteries autonomy Electrical Electrical Electrical reliability of design Electrical Electrical Data hall Busway / cable containment sizing Electrical Electrical IT Power Block Type 1 (MV Distribution Transformer, Diesel Generator, &LV UPS Distribution) Electrical Electrical IT Power Block Type 2 (MV Distribution Transformer, Diesel Generator, &LV Non-UPS Distribution) Electrical Electrical Mechanical Power Block(MV Distribution Transformer, Diesel Generator, &LV UPS &Non-UPS Distribution) Electrical Electrical PDU redundancy level Electrical Electrical Final Whips to racks Electrical Electrical Batteries location Electrical Electrical UPS system redundancy Electrical Electrical Critical Power to Meeh Services Electrical Electrical Generator Location Electrical Generator Generator kick off time Electrical Generator Minimum generator run time after transitioning Electrical Generator Generator rating Electrical Generator Open or Closed Transition Electrical Generator Generator Testing Electrical Data Hall Lighting requirements Electrical Corridor Lighting requirements Electrical Technical room Lighting requirements Electrical Admin / Support Lighting requirements Electrical All Emergency lighting requirements Electrical Fiber Fiber Electrical Fiber Fiber entrance point (zero cable volt ZCV) Electrical Fiber POE rooms Electrical Fiber Network rooms dimensions Electrical Fiber Network / IDFs racks within data halls Electrical Fiber Network underground containment (civil works - from ZCVs to POE rooms) Electrical Fiber Network high level containment (from POE rooms to tenant space) Electrical Fiber Network containment between tenant spaces data halls / offices (single tenant building or more than one DH leased to same tenant) Electrical Fiber Network containment within data hall Electrical Fiber GPS antenna Table 3 - Electrical discipline Table 3 details the various electrical infrastructure considerations for adaptability to diverse client needs. These include distribution systems which can be designed to accommodate varying load densities, distribution panel configurations, and scalable UPS and generator requirements, these include but are not limited to Power distribution units (PDU) - These distribute power to the servers and must be positioned to avoid interference from the cooling system. • IDFs (Intermediate Distribution Frames) - Cabinets that house networking equipment like switches, routers, and patch panels. These act as a crucial point for connecting devices. Cooling needs to consider their location and airflow needs. These are sometimes stored in individual rooms. • Switchgear Cubicles - These handle the electrical switching. They also generate heat and need to be factored into the cooling plan. CHscipline Area System / Component Facility Admin / Support Support Facility Admin / Support Office requirements Facility Admin / Support Other rooms requirements Facility Admin / Support Loading bay requirements Facility Admin / Support Elevators Table 4 - Facility discipline Table 4 details different facility support areas required for a hyperscale data centre, for example the amount of storage space required or how many doors are required for the loading bay areas to facilitate deliveries. Discipline Area System / Component Fire Fire Suppression &Detection Fire Data Hall Data halls Fire Critical Security equipment room Fire Critical Electrical equipment room Fire Generator Generator room enclosure Table 5 - Fire discipline The different areas listed in - Table 5 will have different fire suppression requirements which change depending on the end user, these typically include but are not limited to sprinkler systems, gaseous suppression, levels of smoke and heat detection. The building must be designed to ensure their sufficient space to cater for all these variables. Discipline Area System / Component Mechanical Data Hall Data hall Noise Criterion Mechanical Data Hall Air Supply configuration Mechanical Fuel Fuel System Mechanical Fuel Fuel storage Mechanical Fuel Fuel Bulk Tanks Mechanical Fuel Fuel Pumps Mechanical Fuel Fuel Pipework Mechanical Fuel Fuel control panel Mechanical HVAC Mechanical Critical Target PUE Mechanical Critical Cooling system availability Mechanical Data Hall DH air flow Mechanical Critical Buffer tank Mechanical Critical Make up water tank Mechanical Critical Make up water system redundancy Mechanical Data Hall DH cooling equipment redundancy Mechanical Data Hall Cooling unit requirements Mechanical Data Hall MAU filter class Mechanical Data Hall Humidifier Mechanical Critical Other Critical rooms cooling equipment redundancy Mechanical Critical Critical room ventilation equipment redundancy Mechanical Data Hall External design conditions / DH cooling Mechanical Data Hall External design conditions / DH ventilation Mechanical Admin / Support External design conditions (support cooling and ventilation) Mechanical Data Hall Internal design condition (temperature) Mechanical Data Hall Internal design condition (humidity) Mechanical Data Hall IT racks maximum pressure drop Mechanical Critical Network / electrical rooms internal design conditions (temperature) Mechanical Critical Network / electrical rooms internal design conditions (humidity) Mechanical Critical IDF, POE and secured storage areas internal design conditions (temperature) Mechanical Critical IDF, POE and secured storage areas internal design conditions (humidity) Mechanical Critical Battery room internal design conditions Mechanical Critical Critical spaces differential pressure including DH, network room etc Table 6 - Mechanical discipline Table 6 shows, requirements to ensure the data centre's mechanical infrastructure which can adapt to diverse client needs, key design considerations are prioritised. Data hall environment is designed to accommodate varying noise criteria, adaptable air supply configurations, and scalable cooling unit requirements, with redundancy in cooling and make-up water systems ensuring robust operational resilience. Customisable fuel systems are developed through an evaluation of storage, delivery, and control, facilitating tailored configurations for diverse power backup demands. Resilient critical infrastructure was established, with target PUE, guaranteed cooling system availability, and precisely controlled environmental conditions in critical rooms designed for maximum flexibility and redundancy. Comprehensive environmental control was achieved by meticulously considering external and internal design conditions, including temperature, humidity, and airflow, across all areas, ensuring optimal performance and environmental stability for all potential clients. Preferably, a plurality of air supply temperature in the range 27-35°C, a plurality of room pressure values in the range 0 to 15Pa and a plurality of relative humidity values in the range of 5-80%, for each of the various mechanical, electrical and plumbing configurations in the plurality of electrical rooms, data halls, were used as the environmental controls, Discipline OBiiiiii® System / Component Security / Access Masterplan Security Fence Security / Access External Proximity to security fence Data Hall Data Hall design Data Hall Detection System General Detection System Critical Unit substation enclosure Table 7 - Security access and other disciplines Table 7 outlines the security system infrastructure considerations for adaptability to diverse client needs. The security system design is intrinsically linked to the data centre's physical layout, The additional work above was carried out in evaluating the optimal combinations of key components to ensure that the most flexible and adaptable data centre design could be achieved. Table 8 below shows an embodiment of the method as described, applied to the outcome of a flexible hyperscale design for 2 scenarios, Scenario 1 and Scenario 2, it shows a range of options for a 48MW data centre along with the required quantity of Generators (Engines) based upon alternative options. Table 8, Example of Flexible Data Centre Combinations, Suitable for the Same Building Footprint, Shell &Core and Data Hall. Building with Evaporative 48 MW Building with Air Cooled Chillers 48 MW Building with Evaporative Cooling -Customer at BSdegC (Scenario 48 MW Building with Evaporative Cooling-Customer with in-rack UPS (Scenario 2) Combined 28x2.8 MW 28x3 MW 28x2.6 MW 28x2.6 MW Lineup Arrangement engines (78.4 MW) engines (84 MW) engines (72.8 MW) engines (72.8 MW) Separated 32 x 2.4 MW 34 x 2.4 31 x 2.4 MW 31x2.4 MW Mechanical and IT Lineups engines (76.8 MW) MW engines (81.6 MW) engines (74.4 MW) engines (74.4 MW) Table 8 - Flexible Data Centre Combinations Figures 4 and 5 are for the same data hall Floorplan but at two different IT densities. This invention approach to hyperscale design is a flexible cold-shell able to cater for Scenario 1 and Scenario 2, with the following normal constraints made workable for all designs: Below 5 criteria shows the worst case 1. Density: based on Scenario 1. 2. Column grid: based on Scenario 1 which requires the most racks to utilise the data hall total load, 3. Temperature: base chiller capacity for space planning with flexibility to optimise (increase to Scenario 2 and Scenario 1), 4. Weight: based on Scenario 1 I Scenario 2 - the structure based spread out Scenario 2 5. Electrical Plant: UPS topology for Scenario 1. Additional to these all above, to achieve better Power Usage Effectiveness (PUE) and total energy consumption of the data centre, the following items were looked at overall in deriving this methodology proposed in this invention: • Flexible Cooling Solutions: Different cooling solutions can have a big impact on PUE, such as changing between air-cooled chillers vs evaporative cooling vs VRF. The best cooling equipment depends on a variety of factors such as environmental conditions, e.g. air-cooled chillers are more efficient in colder climates especially when equipped with free cooling circuits whereas evaporative cooling is most effective in hot and dry climates with low humidity. • Optimised Cooling Temperatures: Pushing the cooling temperatures as high as possible within data halls is a key part of the research. The higher the temperatures can go, the lower the PUE will be. This must be done whilst not exceeding the server inlet temperatures which are unique to each potential end user. The higher the room temperature the less the air needs to be cooled meaning less electricity consumption which will have a benefit on the PUE. In an embodiment, the air supply temperature was between 27-35°C. • All cooling systems are based on worst case client IT demands, and system resilience is based around this. The basis of this invention is on the methodology of the design process of a flexible hyperscale data centre design. Although there is a lot of publicly available information and material to follow when designing a typical data centre, there is not a clear guide on designing hyperscale data centres. Therefore, this present invention looked into developing novel methodology when approaching flexible hyperscale data centre design and the various matrices which should be considered. The invention addresses the challenges of balancing the unknowns of the end-user against energy efficiency, cost and different redundancy requirements, this is achieved whilst maintaining a building which can be adaptable at any stage of the design process. This invention can then be applied across different regions, climate zones, and regulatory environments. Finally, proposed methodology can be used as benchmark in hyperscale design development process. Although the present invention has been described in connection with some example embodiments, it is not intended to be limited to the specific form set forth herein. Rather, the scope of the present invention is limited only by the accompanying claims. Additionally, although a feature may appear to be described in connection with embodiments, one skilled in the art would recognize that various features of the described embodiments may be combined in accordance with the invention. In the claims, the term ‘comprising’ does not exclude the presence of other elements or steps. Furthermore, although individually listed, a plurality of means, elements or method steps may be implemented by, for example, a single unit or processor. Additionally, although individual features may be included in different claims, these may possibly be advantageously combined, and the inclusion in different claims does not imply that a combination of features is not feasible and / or advantageous. Also, the inclusion of a feature in one category of claims does not imply a limitation to this category but rather indicates that the feature is equally applicable to other claim categories, as appropriate. Furthermore, the order of features in the claims does not imply any specific order in which the features must be performed and in particular the order of individual steps in a method claim does not imply that the steps must be performed in this order. Rather, the steps may be performed in any suitable order. In addition, singular references do not exclude a plurality. Thus, references to ‘a’, ‘an’, ‘first’, ‘second’, etc. do not preclude a plurality. Acronyms: BoD - Basis of Design CRAG - Computer Room Air Conditioning CRAH - Computer Room Air Handler CHW-Chilled Water FWU - Fan Wall Units HV - High Voltage HVAC - Heating, Ventilation &Air-conditioning IT - Information Technology LV - Low Voltage MEP - Mechanical, Electrical, and Plumbing MUPS - Mechanical Uninterruptible Power Supplies MV - Medium Voltage PDU - Power Distribution Unit PUE - Power Usage Effectiveness SoQ - Statement of Qualification SR - Server Row STS - Static Transfer Switches UPS - Uninterruptible Power Supplies ZCV - Zero Cable Voltage CSA - Civil, Structural &Architectural

Images