This article is the second in a series of Data Centre Best Practice articles provided by Future-tech Ltd.
In this article we seek to highlight the starting points for any data centre project and lay down the basics for success in both the Build and Use phases.
General Planning and Design Best Practices
Before commencing any data centre project it is vital that the purpose of the data centre is fully understood in terms of supporting the business and mitigating any potential operational risks faced by the business. The purpose of any data centre is to accommodate the IT equipment supporting business services and to maintain the level of service availability suitable for that business.
Not all businesses will require high availability or fault tolerant services so the additional cost of obtaining these will be a waste. Conversely a data centre that cannot support the business expectations or mitigate potential risks to achieve the appropriate level of business continuity is ultimately not fit for purpose. It is apparent that this step is missed in many data centre projects with decisions being made about a particular data centre design or solution without fully understanding the implications of that design or the potential business impact it may have.
It is vital to ensure that IT and Facilities infrastructure is appropriate for the specific business risk profile rather than a generic level of resilience. For this to happen it is essential that an integrated IT and Facilities culture within the business is in place from the outset and that decision are made jointly.
Following this it is key to ensuring that business expectations in terms of risk and expected mitigations are properly translated into the site infrastructure and operational platforms.
General Design and Management Principles
The following section details general practices should be considered as a matter of course in any data centre project.
Minimise construction and operating costs by introducing energy optimization at the earliest phases of design; avoid excessive/redundant “safety margins” and right-size to trim first costs.
Right-sizing power and cooling infrastructure is one of the most effective ways to reduce capital and operational costs in the data centre. Work to understand lifecycle requirements and size infrastructure accordingly. Track vendor innovations, and, whenever possible, move toward more modular, flexible, and standardised solutions that improve agility and scalability.
It is generally worthwhile to spend more for infrastructure components that run efficiently at anticipated loads. Power loss in uninterruptible power supplies, power distribution units, cooling systems, etc., merely add to thermal load. In summary try to avoid low partial loads on all infrastructure equipment.
All Electrical and Mechanical systems should be designed and installed with the ability to easily operate, maintain and replace without impact during site operation. Sufficient space and access should be provided for risk free maintenance activities and to facilitate delivery and installation or replacement of large components.
Adequate provision should be made for the receipt of new equipment, removal of packaging, storage of equipment and documentation, assembly and testing. There is often insufficient space for these activities in many data centres with the result that the technical floor space is used inappropriately.
Additionally make sure that floor loadings, doorways and goods lifts are suitably sized to accommodate the equipment anticipated to be installed. This is particularly true for pre-installed cabinets such as OCP deployments which may have individual cabinet weights of over 1 tonne/ton.
Include integrated monitoring, measuring and environmental controls in the facility design. Sufficient space should also be made available for BMS/BAS monitoring and operations, IT operations and control centre, maintenance equipment and tool storage, critical spares, engineering workshop, meeting and training rooms.
Do not share cooling or power delivery systems with other (non-critical areas of buildings. Office areas should always be on separate systems to ensure that both reliability and efficiency of the critical areas are not compromised.
Select and deploy mechanical and electrical equipment which does not require mechanical cooling. The site design, cooling system operational set-points and IT equipment environmental control range should allow the data centre to operate without mechanical cooling for the maximum amount of time practical for the location.
Consider the use of Power Factor corrected mechanical equipment to improve efficiency, however it should be noted that IT equipment can have a leading power factor so there may be a net cancelling effect across the site if properly planned.
Consideration should be given to the provision of connection points, tap offs, supports for future infrastructure equipment installations in order to increase capacity or site load.
Minimise the impact of insolation on the data centre by utilising effective insulation and suitable wall and roof coverings. Do not have external windows in the data centre.
Light coloured equipment cabinets should always be used to reduce the amount of lighting required and minimise energy costs. Always use Computational Fluid Dynamic (CFD) modelling to design data centre technical space and validate IT deployments prior to construction or major refurbishment. The outside of the building should also ideally be modelled to make sure that exhaust from generators does not enter the building and that any heat rejected from the building does not build up around cooling equipment impede the operation of the cooling systems. The overhead and work involved at this stage in minimal compared to construction costs can pay big dividends.
The use of CFD during operation post deployment is expensive and is of less value compared to cost in an operational environment. Thermal Mapping and analysis is more appropriate and cost effective during live operation.
Reference The EU Code of Conduct on Data Centre Energy Efficiency for further information on Energy Efficiency Best Practices (This document is free to download) Data Centres Code of Conduct.
Future-tech have been designing, building and managing business critical data centres since 1982. The experience gained in being involved in the data centre sector from the outset has resulted in Future-tech sites achieved 99.999% uptime during 35+ years of operation. Future-tech has a team of experienced, skilled and highly trained in-house Data Centre Engineers capable of properly maintaining and operating business critical data centre sites of all sizes. For more details please contact Richard Stacey on 0845 900 0127 or at rstacey@future-tech.co.uk