Search this website:

Conditions Change. Adapt Faster. Learn more > Conditions Change. Adapt Faster. Learn more >

Having the measure of data centre efficiency

Ongoing measurement is the only real approach to sustained energy reduction, writes Mike West, Managing Director, Keysource.


Date: 20150905

There is an increasing demand from businesses wanting to gain a greater insight into the performance of their data centre infrastructure and associated operating costs, so that they can be managed more closely and more responsibly. Ongoing measurement is the key and there is no question that this is the only approach to sustained energy reduction. To make any corrective investments it is first necessary to understand where a facility is underperforming and identify what levels of waste are occurring. 

Many businesses need help and advice on achieving efficient power utilisation that doesn't necessarily involve the construction of new facilities, and that introduces more efficient cooling methods to tackle the increasingly powerful, high-density IT servers they are required to install. What is also apparent is that the cooling function within a facility provides the widest scope for energy reduction, so knowing where best to utilise best practice procedures, designs and technologies can dramatically optimise a data centre and achieve considerable operating efficiencies.

The Green Grid has brought together the data centre industry to establish metrics that measure the efficiency of facilities. The key objective is eventually to establish one overall measure for data centre efficiency that includes the physical infrastructure (primarily power and cooling) as well as the IT layer (servers, storage, communication and networking). This measure could, in simple terms, be defined as ‘useful work over power consumption'.

As a starting point, the data centre industry has focussed on DCiE (Data Centre Infrastructure Efficiency) and PUE (Power Usage Effectiveness), which can be easily calculated using standard user management tools. The adoption of these industry standard measurements will make it easier to pursue three clear main energy efficiency goals which are: to minimise the overall power needs of the data centre; to maximise the percentage of the power coming in that is used for IT computing work; and to minimise the amount of power spent on non-IT computing equipment such as power conversion and cooling.

Last year Keysource launched the Data Centre infrastructure Efficiency (DCiE) Assessment, a new affordable service for businesses wishing to save money and minimise environmental impact by reducing the energy consumption of their IT facilities. This evaluation provides a review of a data centre's energy use and is an important first step in the measurement of its Power Usage Effectiveness (PUE).

As part of the assessment, an initial visit is conducted to take the necessary measurements of a facility. This information is then analysed to establish a snapshot of the PUE and benchmarked against the Green Grid's best practice guidelines. Recommendations will then be made targeting design and operational factors that are impacting on the energy efficiency. The DCiE Assessment provides insight into energy saving opportunities and how to lower the total cost of ownership. This advice can help businesses to work towards efficient power utilisation using efficient cooling methods.

For any business looking to reduce energy consumption they must first establish a ‘baseline and track performance'. The DCiE/PUE metric is a great starting point and easy to understand.  However, to truly uncover the potential savings and to establish a robust energy reduction programme, it is essential to drill down further into the infrastructure. This will provide the insight needed to undertake a fuller assessment that takes into account ambient conditions and seasonal effect.

Industry figures suggest that the average PUE of a data centre is currently around 2.5, but the reality is that there is still a large number as high as 7 or even 8 PUE. With a better understanding these underperforming data centres can be fine-tuned or upgraded, but without proper measurement businesses are blind to waste in terms of operating costs, energy usage and emissions.

It has become clear that ‘rightsizing' and the ‘level of redundancy' can have a significant impact on the efficiency of data centre facilities, so need to be taken into consideration. There is also a common industry realisation that some energy losses are quite clearly due to poorly configured temperature and humidity controls and set points, with many data centre cooling systems simply inefficiently deployed and not operated at recommended conditions.

If a data centre has been growing rapidly during the past few years, chances are some servers have been installed wherever there is room, without any underlying plan. This can lead to isolated temperature hot spots, and an overall environment that is exceptionally hard for cooling systems of any type to control efficiently. Rearranging the physical position of servers in a data centre can significantly reduce the amount of overall cooling capacity required, although the total amount of heat generated remains the same.

Moreover, it is vital that all the key elements that affect how heat is reduced and removed from the facility should be considered as a whole in order to achieve maximum efficiency. Hot and cold aisle layout with blanking plates is very effective, but it's the control of the warm return air and minimising hot and cold air mixing that is truly critical to avoiding hot spots, maintaining ‘redundancy' and reducing energy.

The design of modern data centres has become increasingly sophisticated by applying complex software like Computational Fluid Dynamics (CFD) and airflow modelling.  Using these powerful 3-D software solutions to recreate a facility in the virtual world, from the walls and floor to the racks and equipment in them, provides a graphic analysis of how hot and cool air flows from rack to row and from row to room. This enables enhanced planning of cooling infrastructures for new builds and addresses existing data that have heat issues.

More detailed analysis of the data centre space is clearly required, but progress is being made. The data centre industry needs to take the lead and be at the forefront of rolling out ‘best practice' energy efficiency assessments to businesses and incorporating these metrics into the design of new facilities. This will provide a road map to running a more efficient facility and ultimately result in far wider benefits than just reducing the energy consumption of a data centre. Other benefits will include lowering the total cost of ownership, improving the overall performance of a facility and solving any hot spot problems.

In conclusion, ongoing measurement is the only real approach to sustained energy reduction. The good news is that rising power costs can represent 25 per cent of the total cost of data centre ownership, so any efficiency improvements can have a significant positive impact on overheads and carbon footprint. 

Surrey that would significantly reduce power usage to achieve considerable cost savings and reduced carbon footprint. Keysource was appointed to design, build, install and manage the facility following a competitive tender.


PGS offers a broad range of products to the oil industry including; seismic and electromagnetic services, data acquisition, processing, reservoir analysis/interpretation and multi-client library data. The company helps oil companies to find oil and gas reserves worldwide, offshore and onshore.


. The facility has been designed for 1.8MW of IT load and can accommodate high-density IT hardware at any rack position with rack power densities of 20kW comfortably supported, well within the 15kW per rack specified by the PGS IT team.


Initial results confirm that the data centre has achieved a Power Usage Effectiveness (PUE) ratio – an energy calculation recommended by industry sponsored organisation Green Grid – of 1.2 compared to a typical UK figure for a conventionally designed data centre of 2.2. Therefore, at full IT load and compared to their previous site the Ecofris cooling solution will reduce annual power consumption by 15.8 million kW-h and reduce emissions by 6.8 million tonnes of CO2 per year (Based on UK grid average). Furthermore, the facility has a DCiE of 80%, where 80% of the total power being consumed by the facility is powering the IT equipment.

Keysource completed all the detailed design for the facility including a new mains supply transformer, N+1 power and cooling solutions as well as supporting protection systems. A modular approach has been utilised to ensure seamless upgrade from the 600kW phase 1 IT load through to full deployment.

Central to the design process, and key to the new concept, was the use of CFD (Computational Fluid Dynamics) software to model the data centre airflow. Keysource's in-house experts have used CFD modelling extensively in the development of the Ecofris cooling solution to ensure that the rack layout and server airflow requirements are optimally matched with the air handling equipment to maximise efficiency and assure the environmental conditions are maintained under all scenarios.

The data centre build was managed by the Keysource's project management team which delivered the project on-time and within budget, handing over the facility with server racks powered up and all network cabling in place to allow the PGS IT team to concentrate on populating the racks with the servers.

During the closing stages of construction comprehensive system testing was carried out to verify that the facility operated within the design parameters, with Keysource taking a strong hand in the final system commissioning to make sure they had a comprehensive understanding of the new systems they will be maintaining, and simulating all conceivable scenarios to make sure all the plant and equipment will work reliably, as one system, once in operation.

A comprehensive management and monitoring solution has been installed to alert the Keysource Facility Management team directly in the event of any system alarms. The system brings all the data centre plant and equipment together in one view to simplify ongoing management of the space. This provides high level management reports to track the capacity and performance of the facility, with PUE constantly recorded and reported as one of the Key Performance Indicators.


Mike Turff, Global Datacentre Manager at Petroleum Geo-Services says: "We are committed to reducing energy, but we also needed to ensure that any data centre design would allow deployment of our high-density server and storage hardware. The safe and efficient storage of vast amounts of vital data is key to the success of our business, however we are also highly committed to a strong environmental stance. The Keysource Ecofris solution provides us with a resilient and future proof answer without compromising our core values."


« Previous article

Next article »

Tags: Green Tech, Power & Cooling

More Exclusive News

Intelligent IT solutions for the Internet of Things

31 Aug 2015
By Patrick Steiner, Lead Architect at Red Hat.

Top reasons to shift to a managed network services approach  

31 Aug 2015
By Sean Paxton, product manager, Redcentric.

How data centre security can uphold online democracy

31 Aug 2015
By Paul Dignan, Field Systems Engineer, F5 Networks

The case for cloud data backup is growing stronger  

31 Aug 2015
By Wieland Alge, General Manager & VP EMEA at Barracuda Networks.

How to maximise Cloud efficiencies in the data centre using people, process and technology

31 Aug 2015
By Gerardo Dada, SolarWinds.

Scheduled downtime: Minimise the pain by making it rare  

31 Aug 2015
By Doug Hazelman, VP Product Strategy, Veeam.

Put information security ahead of infrastructure security  

24 Aug 2015
By Mark Edge, Country Manager UK, Brainloop.

Managing capacity in a world of relentless data generation

24 Aug 2015
By Nigel Stevens, Managing Director for IO in the United Kingdom.

Clearing the way to cloud services – five tips for a safer journey to the cloud  

24 Aug 2015
By Chris Pace, Head of Product Marketing, Wallix.

Understanding the difference: Disaster Recovery and Data Backup

24 Aug 2015
Backup Software and Disaster Recovery are often mistakenly placed in the same category. Harking to an early era when enterprises tended to run batch processing systems and the working day was 9 to ...

Adaptive storage: compromise or the best of both worlds?

24 Aug 2015
Gavin McLaughlin, VP of Strategy and Communications at X-IO technologies. 

Integrating IT Risk Management with the business  

24 Aug 2015
Businesses are now so reliant on technology that many simply couldn’t operate without it. IT has become such an integral part of business that IT risk management is an imperative for all orga...

Unappy times for SMB password management  

17 Aug 2015
By Boris Jabes, Senior Director of Product Management at LogMeIn.

The search for 'best in class'

17 Aug 2015
Nimbus Hosting is a UK based managed hosting company which provides high quality Linux and Windows cloud web hosting solutions for small and medium sized businesses. The company wanted to find ...

Intelligent IT solutions for the Internet of Things  

17 Aug 2015
By Patrick Steiner, Lead Architect at Red Hat.


Latest IT jobs from leading companies.


Click here for full listings»