Mission Critical Power by The Energyst

missioncriticalpower.uk

ISSUE 18: August 2018

Rise of the robots and the global war on talent: what does the future hold for data centre professionals and will they be replaced by machines?

Hyperscaleinspired Open Compute Project principles are being applied to the latest data centre builds

How will data centre technology adapt to the demands of the Industrial Internet of Things and ensure resilience at the industrial edge?

Protecting your assets: the importance of battery testing See page 14

IN THIS ISSUE

12 Leadership What qualities will be required to lead data centres in the future in an era of rapid change?

Data centre management

Optimisation Hyperscale inspired: the latest data centres are being built to be OCP ready

Will data centre professionals be replaced by robots?

30 Power management A major hospital in the US has reduced its energy costs. Could UK hospitals follow suit?

missioncriticalpower.uk

ISSUE 18: August 2018

Rise of the robots and the global war on talent: what does the future hold for data centre professionals and will they be replaced by machines?

Demand-side response

Hyperscaleinspired Open Compute Project principles are being applied to the latest data centre builds

How will data centre technology adapt to the demands of the Industrial Internet of Things and ensure resilience at the industrial edge?

Protecting your assets: the importance of battery testing

Irish Water and Northern Ireland Water are paving the way for DSR

See page 14

Industrial IoT

Front cover

How can you reduce the risk of unplanned downtime at the industrial edge?

Meggar discusses the need for battery testing to ensure the resilience of mission critical sites

Comment

UPS

Demand-side response

News

Thermal Management

Cables & Connections

Data Centre Management

Power Distribution

Products

Back-up Power

Q&A

Viewpoint

To subscribe please contact: missioncriticalpower.uk/subscribe missioncriticalpower.uk

August 2018 MCP

COMMENT

Giving UK industry the edge There is a lot of discussion on the ‘Internet of Things’, but one area where digitisation is having a dramatic impact is in the industrial sector. GE estimates that the Industrial Internet of Things (IIoT) could be a $225bn (£172bn) market by 2020. There is huge potential for IT technologies to increase efficiency, reduce energy costs and ensure reliability for critical operations. IIoT combines machine-to-machine communication, industrial big data analytics, technology, cyber security, and HMI and Scada, to drive high levels of productivity and performance. Industrial sites today want more and more data and analytics but latency is a key consideration and many applications need to be operated at the edge. This trend is seeing the implementation of robust infrastructure and technologies designed specifically to support the placement of more compute and storage capacity closer to industrial environments. Victor Avelar says that Schneider Electric is seeing more on-premises IT kit, driven by an upsurge in IIoT. In particular, servers are being placed in micro data centres. According to predictions by 451 Research (2018 Trends in Datacentres & Critical Infrastructure), this need for localised capacity, means that micro-modular data centres (MMDCs) will become a key component of future IT infrastructures.

As with many critical sectors, uptime is crucial for production lines where millions of pounds can be lost in the event of unplanned downtime. Furthermore, systems are often safety critical, as well as business critical. Resilience has become part of the DNA of data centres, and now the industrial sector is starting to catch up, placing increasing demands on the infrastructure. As digitised systems grow in popularity across industries, uptime in the form of steady consistent power protection and cooling will become ever more critical to business productivity and profitability. According to a report published by the Royal Academy of Engineering, titled ‘Counting the cost: the economic and social costs of electricity shortfalls in the UK’, energy security issues are climbing up the industry agenda in the UK:

“...Many industrial and commercial processes are increasingly reliant on a continuous, high-quality power supply, especially where just-in-time supply chains are used. Because of this, interruptions or voltage sags for even a couple of seconds could have an extremely detrimental impact upon operations…” Jason Andersen from Stratus Technologies predicts that with the increasing adoption of edge-based systems, including IIoT technologies, the costs and risks associated with downtime will continue to trend upward. As manufacturing processes become more reliant on IT, the resilience of the hyperscale data centre will need to be adapted and adopted for the industrial edge. Louise Frampton, editor

Editor Louise Frampton louise@energystmedia.com t: 020 34092043 m: 07824317819 Managing Editor Tim McManan-Smith tim@energystmedia.com Design and production Paul Lindsell production@energystmedia.com m: 07790 434813

MCP August 2018

Sales director Steve Swaine steve@energystmedia.com t: 020 3714 4451 m: 07818 574300

Energyst Media Ltd, PO BOX 420, Reigate, Surrey RH2 2DU Registered in England & Wales – 8667229 Registered at Stationers Hall – ISSN 0964 8321 Printed by Warners (Midlands) plc

Commercial manager Daniel Coyne T: 02037517863 M: 07557109476 E: daniel@energystmedia.com

No part of this publication may be reproduced without the written permission of the publishers. The opinions expressed in this publication are not necessarily those of the publishers. Mission Critical Power is a controlled circulation magazine available to selected professionals interested in energy, who fall within the publishers terms of control. For those outside of these terms, annual subscriptions is £60 including postage in the UK. For all subscriptions outside the UK the annual subscription is £120 including postage.

Circulation enquiries circulation@energystmedia.com

missioncriticalpower.uk

NEWS & COMMENT

Constraint trading, grid access and Triad: Ofgem scopes charging review Ofgem has broken cover over major changes to network charging and access. The regulator is trying to ensure that the way businesses and households are charged for using the grid is reflective of the costs their actions cause. It admits that distributed generators are exploiting loopholes in the current arrangements, and wants to force “developers to respond to signals that reflect the economic reality, not the peculiarities of the regulations”. Ofgem therefore plans a Significant Code Review to redesign charging and access arrangements, and needs to finish the redesign to coincide with the next set of price controls for network operators, in 2022/23. The regulator indicated it aims to better define access rights and choices within the review. However, it will likely leave some elements of access reform to the networks themselves, such as continuing to providing different sorts of flexible connections, and enabling companies to trade

Distributed generators are exploiting loopholes

curtailment and constraints with one another where they might ordinarily have to stop exporting generation. Letting industry deliver some aspects of change would allow the regulator to concentrate on core reforms to transmission and distribution charging. DUoS Ofgem said it may look at

creating a new zonal charging structure for distributed generation. At present, distribution system charges (DUoS) treat generation as negative demand. So generators tend to get paid, even if their generation is not beneficial to the system. That may change, with Ofgem mooting zones that are demand-heavy or generation

heavy, with distributed generation charged – or credited – accordingly. Triad/TNUoS Ofgem dropped further hints about killing off the current Triad system within the consultation. The regulator said that while the Triad regime has merit, it may also be introducing uncertainty, as predicting Triad periods becomes increasingly difficult, and causing distortions due to the differences in the way small and large generators are treated and charged. For generators, Ofgem may move to treat charges for both distributed generation and transmission in the same manner, which could see distributed generators lose another chunk of income. For demand customers, Ofgem said options could include moving away from charging based on demand during Triad periods to fixed time of use windows, such as the bands used for DUoS charges.

County Durham colo announces site expansion Critical environment specialist Keysource has upgraded and expanded the colocation facility of managed IT solutions and services expert ITPS in County Durham. The need for the expansion was driven by increased demand for ITPS’ services, following its selection by Microsoft as one of the early adopters of the Azure Stack roll-out. The 30,000 sq ft site has a total 1.6MW capacity, with 900kW partitioned into different data halls for specific clients. The upgrade added an additional 300kW, leaving ITPS with further spare capacity to expand in the future. Keysource completed the upgrade

MCP August 2018

to the second hall at the 8,500 sq ft data centre in just 12 weeks. The company originally designed and built the site in 2014 to act as the flagship hub of ITPS’ network operations, which spans three data centres across the UK. A bespoke modular design was used, which enables high levels of efficiency and resilience, and allows capacity upgrades to be delivered in line with customer demand without disrupting existing operations. The data centre was fundamental in making ITPS one of only a handful of worldwide ICT providers to be selected to join the Microsoft Azure Stack Early Adopters programme.

It is also the only North East stage two-certified supplier for NHS Digital, delivering services across the new UK-wide NHS communications network, the Health and Social Care Network. Garry Sheriff, managing director at ITPS, said: “We’ve seen substantial growth over the past three years and, thanks to a series of new client wins, we’re expanding our capacity much sooner than expected. Our decision to create a highspecification data centre here in the region was based on predicting the growth in Cloud services and a shift towards effectively renting, rather than buying, ICT infrastructure.”

missioncriticalpower.uk

7 OCP to boost liquid cooling innovation The Open Compute Project (OCP) has launched an Advanced Cooling Solutions Sub Project in response to an industry need to collaborate on liquid cooling and other advanced cooling approaches. The project aims to: 1) harmonise the supply chain to enable wider, quicker, and easier adoption of liquid cooled servers, storage, and networking gear; and 2) to create a community of collaborators for future advanced cooling solutions. This new sub-project will be part of the existing Rack and Power Project. High-performance computing has used liquid cooling for many years. These solutions have proven to be costeffective and efficient at heat extraction. With the growth of 5G, IoT, VR, CDN and latent sensitive applications, data centres are being constructed closer to their customers and often in regions

of the globe where traditional air and mechanical (eg chillers) cooling becomes quite expensive. Increased power density also introduces cooling challenges. In these cases liquid cooling, and specifically warm water cooling, becomes an effective alternative for heat extraction. The new sub project will focus on standardisation and definition of critical interfaces, operational parameters and environmental conditions that enables a non-proprietary, multi-vendor supply chain for ‘warm water’ cooling. OCP envisions a supply chain offering a variety of IT devices (servers, storage, networking etc) that can work with a variety of liquid-enabled racks from many solutions providers. Direct contact, immersion and other advanced cooling options are within the scope of this project.

Sudlows wins major contract with global telecoms provider

Critical infrastructure specialist Sudlows has secured a multimillion-pound infrastructure contract with global communications provider Telefónica UK. Telefónica is a provider of mobile services, offering communications solutions to consumer and corporate customers across the globe. The project involves the design and installation of a missioncriticalpower.uk

system upgrade of Telefónica UK’s existing UPS systems with a range of modern UPS solutions and associated electrical switchgear. This critical infrastructure project will also provide enhanced cooling capability to support the UPS operation, delivering N+N resilience and substantial operational energy savings. Sudlows' commercial director, Chris Dummett,

commented: “Sudlows has been at the forefront of completing complex critical infrastructure projects, in the UK, over the last decade and this latest contract with Telefónica UK will see the company deliver a major replacement programme of resilient and energy efficient modern UPS systems throughout the Telefónica UK estate.”

News in brief Construction starts on Frankfurt data centre Global Switch has started construction on its second data centre in Frankfurt. The new state-of-the-art facility, Frankfurt North, immediately adjacent to the existing Frankfurt South data centre, will provide around 11,000 sq m of gross space with 14MVA of utility power, bringing the total power capacity of the campus to 28MVA. The facility will be targeting a Breeam “Very Good” certification and is due to launch in 2019. ABB acquires GE brand ABB has completed its acquisition of GE Industrial Solutions (Geis), GE’s global electrification solutions business. Geis will be integrated into ABB’s Electrification Products (EP) division, led by EP president Tarak Mehta, as a new business unit called Electrification Products Industrial Solutions. Stephanie Mains, who was previously the president and CEO of the GE Industrial Solutions business, will lead the new business unit. Interxion opens third London data centre Interxion has opened its third data centre at its London campus. The new data centre is aimed primarily at businesses in the financial sector, advertising and media industries and offers 1,800 sq m of white space and 3MW of power capacity. The LON3 data centre offers an SLA of 99.999% power uptime, and features N+1 UPS and dieseldriven generators; 13MVA of power (24MVA reserved) and high-density power availability. Interconnection opportunities include direct access to major Cloud platforms, BT’s IP network and the UK’s largest Internet Exchange, Linx.

August 2018 MCP

DATA CENTRE MANAGEMENT

Rise of the robots and the global war on talent The data centre industry has changed at a rapid pace in the past 10 years and will continue to change during the next decade. But what will this mean for data centre professionals? Will they be replaced by robots and will they still be working in large data centres? What will the sector look like in 10 years’ time and what skills will be required? Louise Frampton reports

BRE Data Centre Solutions has grown 50-60% year-on-year and currently has more than 6,000 data centre technicians, employed across 800 data centres worldwide. As the data centre sector continues to grow exponentially, recruitment is becoming highly competitive and there is a “global war on talent”, according to CBRE’s COO, John Dunsten. This is presenting the sector with challenges, as well as opening up opportunities for skilled professionals. Speaking at Anglia Ruskin University, at a ‘Tech Leaders’

MCP August 2018

Address’, Dunsten commented that data centre leaders must be adept at managing change, while embracing technology that can enable data centres to operate more efficiently – including augmented reality and artificial intelligence. Qualities, other than traditional technical skills, are becoming increasingly important and staff will need to be multi-skilled. “Ten years ago, the typical facility was on premises for enterprise end users, resilience was built into the M&E structure, it was hugely complex and there was a race to build the most resilient data centre in the

world. Sometimes, this actually decreased resilience because of the complexity. Therefore, we needed highly technical staff,” explained Dunsten. He pointed out that there was a clear line between M&E and IT, and they rarely met, while the data centre sector was a very secretive industry. “People didn’t really know about data centres or understand them,” Dunsten continued. From a portfolio perspective, there have been significant changes, he pointed out: “Today, people are maintaining a hybrid approach. There is cross training of technicians to provide missioncriticalpower.uk

converged services across IT, M&E and other projects. This is good news for technicians; this multi-skilled approach has made the profession much more attractive. The industry as a whole is exciting and much more public, today – it is now a talking point at a party; in the past, people didn’t know what you were talking about.” He added that staff now have to be much more focused on end user clients, as reliance on customer service in data centres has increased significantly. “Only a few years ago, the scenario in the industry was: ‘This is what you get and this missioncriticalpower.uk

is what you pay for it, whether you like it or not’. Today, there is so much choice and variety that data centre providers have had to become much more customer focused – data centre professionals need to have skills in this area, therefore,” Dunsten explained, adding that it has also become important to have the ability and desire to ‘learn, change and evolve’. “If they don’t have this desire, they are in the wrong business,” Dunsten continued. Key trends He pointed out that the emphasis is now on flexibility

and there is a trend towards non-24/7 coverage – 45% of CBRE’s operation, globally, is now non-24/7. “Five to six years ago this would have been unheard of,” said Dunsten. “We now have clients building big, hyperscale data centres with no power resilience because they take the view that the grid has been up 99.999%. They install a pretty big UPS for a flip, and they are done,” he continued. However, outages have a higher impact on businesses, today, and incidents have become much more public, as there is increasing reliance on data and technology in business and in our everyday lives. A different set of skills is now required, therefore, and leaders must be able to manage this. Nevertheless, organisations such as banks, for example, are building resilience into their software layer. CBRE believes that this will continue to be an important trend. He also highlighted the fact that there is more collaboration across the industry, today: “We have moved out of the shadows and into a more collaborative industry, so there is a need for ‘softer’ business skills for our technicians and leaders, rather than just good technical skills.” Other important trends include the fact that energy management is much more important today, compared to just a few years ago, and knowledge and skills in this area are key to delivering the business objectives of the operation. Robots and AI In the next 10 years, CBRE believes that there will be an increased focus on edge data centres, based around a ‘hub and spoke’ model. This is being fuelled by the need for instant data. CBRE also predicts that 99% of data centres will be ‘non24/7’ in the next decade. “There will be an increasing need for mobile technicians. They will be lone workers but we will still need to ensure they remain connected with the centre of operations,”

There will be an increasing need for mobile technicians. They will be lone workers, but we will still need to ensure they remain connected with the centre of operations commented Dunsten. “We will see the need for deeply technical engineers, located centrally and engaged via technology (possibly augmented reality), with semi-skilled technicians on the ground. They will be able to walk them through procedures and diagnose problems, enabling them to fix any issues. We are investing a huge amount of time and money in making sure we are ahead of the game in this respect,’ he continued. Robotics and artificial intelligence will have a greater role to play in data centres in a very short space of time, according to Dunsten. He revealed that CBRE is working closely with IBM and other partners to this end. “We are already seeing clients building data centres that allow the installation of 9ft high racks. We won’t be carrying step ladders around the data centre. In the future, robots will be employed to pull out servers and replace components. AI is an exciting area for us,” adds Dunsten. At present, engineering talent is being used for ‘checking’ the data centre operations but AI will increase the efficiency of fault finding and monitoring, moving forward, he pointed out. Preparing for the future So how is CBRE preparing for tomorrow’s data centre operation? Dunsten revealed that CBRE is currently exploring possibilities for a central » August 2018 MCP

DATA CENTRE MANAGEMENT

engineering hub, with the top technical leaders from across the world, located in one place, and connected via AR technology to technicians on the ground. They have also hired a dedicated chief reliability and innovation officer, Mike Doolan, who is responsible for looking to the future, where the next rapid change is going to happen, and ensuring CBRE moves quickly and innovates to stay ahead of the game. Most recently, Doolan has been working closely with AI company Litbit. Earlier this year, CBRE and Litbit announced a strategic relationship to transform facilities maintenance within data centres and critical environments. Developing REMI CBRE is training a Litbit AI persona named ‘Remi’ (Risk Exposure Mitigation Intelligence) on the normal operating conditions of its data centre facilities and equipment globally, which can then be used by CBRE’s expert teams for predictive maintenance and rapid problem diagnosis. Remi will run both in the Cloud via smartphone and on the edge via permanently deployed computing devices. It will aggregate the knowledge of CBRE’s worldwide service experience, thousands of technicians, and millions of machines into a single system. It is hoped that the AI solution will provide improved coverage for manned locations and basic coverage in sites that are not currently viable for human staffing. He went on to highlight some of the important considerations for CBRE, moving forward over

We are already seeing clients building data centres that allow the installation of 9ft high racks. We won’t be carrying step ladders around the data centre. In the future, robots will be employed to pull out servers and replace components the next 10 years, including: the need to understand regional variances in data centre maturity models; maintaining flexibility in a rapidly changing environment and embracing technology. “If we stand still for a minute, we will die,” he concluded. Global impact Ultimately, data centres have become the powerhouse of the growing global economy and merged with our everyday lives. Speaking at the keynote address, Alfonso Aranda Arias,

head of global data centre operations, IBM, pointed out that data centres have become part of the “fabric of society”; innovation in science, energy, manufacturing, healthcare and education are all powered by data centres. Arias quoted Mark Weiser, chief scientist at Xerox Parc (widely considered to be the father of ubiquitous computing) who commented that the most profound technologies are “those that disappear. They weave themselves into everyday life until they are

indistinguishable from it.” While data centres have weaved themselves into everyday life, they are now more visible than ever before, as our reliance on instant data is at its most transparent when an outage interrupts normal service. In the past, data centres were discussed at a technical, facilities level but they are now of such strategic importance to global business, discussion has been elevated up the ranks to the top layers of management. “They are key components to the organisation’s journey to value. Data centres require a different kind of mindset. It is not just about M&E and IT. Finance is now part of the equation,” commented Arias, adding that the data centre has become a ‘utility’ component. Data, market intelligence, artificial intelligence and financial management are becoming integrated. He observed that AI is constantly evolving and this means less people on the floor, as sensors and machines are performing the functions once undertaken by humans. These people are now in the central office, interpreting the data and managing contracts. “At the moment, I am not seeing less people,” commented Arias. However, he observed that we are moving out of a ‘microscope phase’ (focused on optimisation) to a mixed environment, where we are also entering a simultaneous ‘telescope phase’ – looking to the future and the next thing. “You need to do both. We are in a hybrid era,” he concluded. l

CNet Training, a provider of technical education for the digital infrastructure industry, hosted the Tech Leaders Keynote Address at Anglia Ruskin University, in Cambridge, with tech giants IBM and CBRE. The address, which was in support of the world’s only Master’s Degree in Data Centre Leadership and Management, explored the key issues associated with data centres being the powerhouse of the growing global economy. Pictured from left to right: Andrew Stevens, CEO, CNet; Dr Terri Simpkin, programme leader and co-author of the Masters Degree in Data Centre Leadership and Management; John Dunsten, chief operating officer, CBRE; and Alfonso Aranda Arias, head of global data centre operations, IBM. MCP August 2018

missioncriticalpower.uk

DATA CENTRE LEADERSHIP

Shaping data centre leaders of the future Andrew Stevens, CNet’s CEO, speaks to Louise Frampton about the qualities required to lead data centres in an era of rapid growth and technological change. He argues that a bridge needs to be built between technical acumen and business leadership skills

he data centre sector president and CEO of CNet. is growing at a rapid New managers and leaders pace and changing will need to understand the beyond recognition. Keeping diversity of the industry, how up with the demands of a such it is changing, and how their a dynamic industry data centre fits into their requires an elite organisation’s eco breed of leaders. system. Data centre “The business managers need outcomes and to be capable the reason for of navigating the data centre’s the challenges existence is of new business key,” Stevens and infrastructure explains. “There models, ensure have been times their teams deliver when organisations against challenging Andrew Stevens have built data centres objectives, while with huge resilience having the technical skills because someone wanted to to embrace technological build the biggest and the best. revolution. At the same “They didn’t actually meet time, they are tasked with the business or conceptual protecting the reputation of requirements of that data their brand, not only through centre; these organisations ensuring faultless service, but are now left with legacy data by meeting tough targets on centres that carry a huge cost sustainability. – once this becomes apparent To meet these challenges, on the balance sheet, they will the industry needs to build want to move on; the model is a bridge between technical going to change dramatically. acumen and the emerging You will need a flexible, mobile demands of business approach to the business as leadership and management, it moves forward,” Stevens says Andrew Stevens, continues.

Developing talent, transferring skills In an extremely competitive recruitment market, individuals with a technical bias are increasingly being promoted into management positions with no experience and they are being challenged. CNet is therefore providing a framework for technical people to gain management skills. Data centre businesses want to retain their staff and one of the ways they are doing this is to support further educational opportunities and personal development, such as the Masters Degree in Data Centre Leadership and Management. The distance learning programme has been developed by CNet, in partnership with Anglia Ruskin University, and the first graduation will take place in October.

The data centre sector, in the past, has been poor at looking outside of its own industry – it wanted to create its own identity, its own processes, its own people and job titles. Now we need to learn from other industries and adopt the best from those sectors MCP August 2018

The course not only targets individuals that are continuing on a career path but also those who wish to migrate into the data centre environment, transferring applicable, traditional skills from other sectors. According to Stevens, the industry needs to look for talent and knowledge beyond the four walls of the data centre. CNet is therefore looking at best practice in other mission critical industries to evaluate where this could be applicable. “The data centre sector, in the past, has been poor at looking outside of its own industry – it wanted to create its own identity, its own processes, its own people and job titles. Now, we need to learn from other industries and adopt the best from those sectors,” comments Stevens. Aviation, transportation, the military and other critical industries have extensive experience of managing disaster situations and developing processes to minimise risk. This knowledge could prove invaluable. People management missioncriticalpower.uk

Individuals with a technical bias are increasingly being promoted into management positions, with no experience, and they are being challenged skills and understanding of human factors will be crucial, moving forward, according to Stevens: “Understanding the single points of failures within the operation is key and often this comes down to people and not the technology. “As an industry, we have moved away from being as technically focused as we once were; the view was that technology would solve everything. “Now we are moving to a point where we understand that people deploy technology and they are the single point of failure. “Whether it is design or maintenance, it is all about understanding the implications and actions,” he explains. Stevens points out that data centre leaders will need to develop the skills to manage risk across multidisciplinary teams, as well as within environments where there may be clashes of culture and differences in technical approaches. “There has been a merger of facilities management and IT, which has been a battleground missioncriticalpower.uk

Managing data centre sustainability Sustainability is increasingly becoming an important aspect of data centre management and CNet has responded by developing a one-year distance learning programme to help develop skills in this area. The in-depth content walks through all stages of data centre sustainability from strategic vision and business drivers, operational analysis of power, cooling and IT hardware, operational processes and procedures, risk evaluation and mitigation, to design innovation and implementing initiatives, while appreciating the business and operational challenges that can be encountered. Maintenance strategies, continuous planning and critical analysis against identified targets and demonstrating ROI are also explored. Successful learners gain Certified Data Centre Sustainability Professional (CDCSP) certification, a post nominal title and a level six Diploma qualification, equivalent to a Bachelors Degree. “A focus on sustainability is a natural progression for CNet to sit at qualification level six on the Global Digital Infrastructure Education Framework. It fully complements the level five programmes that focus on data centre design, management, energy efficiency and auditing. Being such a high-level programme, it is not going to be easy, yet the knowledge data centre professionals will gain as a result will be hugely valuable,” Stevens comments. for many years, but it has to happen. Managers have to lead this movement; they have to be able to manage conflict and they have to be able to manage risk. This includes the application of human factors science. There is a huge role for skills in this area. Organisations that are moving the fastest and are the most successful are adopting human factors learning as a major part of their

operations,” he comments. According to Stevens, the priorities for data centre leaders in the next five years will be to keep up with the pace of growth and be skilled in managing mergers and acquisitions, as well as the migration of different businesses as part of this. “People can be resistant to change, particularly when a new culture is imposed by the acquiring company. We have

to look hard at these areas – a manager may have a team of 15 to manage and, three weeks later, find they have 150 people, from different backgrounds, different skills sets, different cultures and different technical approaches.” Ultimately, leaders will need the skills to bring people on board to do things their way and have the flexibility to navigate change. l August 2018 MCP

Advertorial

Protecting your assets: the importance of battery testing Megger’s Antony Wills discusses the need for battery testing to ensure the resilience of mission critical facilities as sites cannot afford to become complacent

tandby batteries sit unused for 99% of the time and, as the UK has quite a reliable power supply, it is easy to become complacent. Yet battery failure is a leading cause of load loss. There are some myths and misconceptions in the market that need to be dispelled and the quality of testing still varies widely throughout the UK. There is a tendency to fit and forget batteries but it is critical that your assets work when called upon, so testing is essential to prevent any costly downtime. Some people believe that batteries, that are advertised as being ‘maintenance-free’, do not need to be tested, but this simply means that you cannot perform maintenance on the battery – it can still fail before the end of its designed life. Battery

capacity can drop significantly long before its calculated life expectancy is reached, and the battery can have weak cells that can cause costly service interruptions. Capacity testing A reliable way of measuring battery capacity is to conduct a discharge test, which can be done with a series of test instruments that enable sites to perform discharge tests at constant current, constant power or constant resistance, like our Torkel range does; users can even test the batteries on-line and get extra load units if they need a higher load current. Our BVM, a battery voltage measurement device that is used for individual cell voltage measurement of large battery banks, can also be used in conjunction with the Torkel to perform a completely automated battery bank capacity test. Impedance testing To complement this, impedance testers are also available. The battery impedance test helps to identify weak cells before they cause problems. However, taking the battery off-line for testing is time-consuming and risky – but you don’t need to do

this if using the online testing capabilities of Megger’s battery test equipment. For example, the BITE line of battery testers will perform online testing that will determine the health of lead-acid batteries up to 7000 Ah. The BITE range of battery testers measures cell impedance, cell float voltage and inter-cell connection resistance, as well as ripple and float current. Key testing considerations End users need to consider the practicalities of the test. If they are going to use a discharge tester and remove most of the energy from the battery system, they must ensure the operation is not left vulnerable in the event that the battery is called upon towards the end of the discharge. Impedance tests are very quick and give you a good level of confidence that the battery is going to be ok, but it is good to correlate this with discharge tests – at intervals – so that you can decide when to swap out the entire battery. Some sites may have their

own built-in diagnostics and reporting systems. However, in general, it is recommended that batteries are tested at: delivery, to get a baseline value; when they have reached a quarter of their life; half way through; and then at 75% of their life. After this, you will need to conduct more regular testing to verify that the battery will be in good condition towards the end of its life. Even a single cell failure may mean that the battery cannot deliver its power anymore, so it is important to identify if any of the cells are going to fail towards the end of the battery’s life. Ultimately, regular capacity and impedance testing is essential for predictive maintenance and protection of assets, and it must not be time-consuming or expensive. Megger supplies a range of easy to use testers designed to provide either a full test or just a quick check of the battery’s condition. For more information, visit uk.megger.com or email uksales@megger.com

DATA CENTRE OPTIMISATION

penness is not something normally associated with the data centre sector – data centres have historically kept their secrets closely guarded, protected by non-disclosure agreements, and hidden behind anonymous, high-security walls. This could be about to change, however, with the adoption of open compute principles. The Open Compute Foundation believes that sharing ideas, specifications and other intellectual property is the key to maximising innovation. So, what exactly is an ‘OCP-ready’ facility and how can wholesale colocation providers benefit from this expanding market? Project origins In 2009, Facebook was growing exponentially, offering new services and giving millions of people a platform to share photos and videos. Looking ahead, the company realised that it had to rethink its infrastructure to accommodate the huge influx of new people and data, and also control costs and energy consumption. This prompted Facebook to embark on the challenge of designing the world’s most energy efficient data centre; one that could handle unprecedented scale at the lowest possible cost. A team of engineers spent the next two years designing and building from the ground up: software, servers, racks, power supplies and cooling. The resulting data centre, which now stands in Prineville, Oregon, was 38% more energy efficient to build and 24% less expensive to run than the company’s previous facility. In 2011, Facebook shared its designs with the public and, along with Intel and Rackspace, Goldman Sachs and Andy Bechtolsheim, launched the Open Compute Project. The five members of the OCP Foundation hoped to create a movement in the hardware space that would bring about the same kind of creativity and collaboration seen in open source software.

MCP August 2018

Hyperscale inspired: getting ready for OCP The Open Compute Project (OCP) is seeking to adapt hyperscale innovation and roll this out across the data centre sector. Sharing efficiency and cost-saving approaches across the sector has the potential to advance practice and transform the industry. Kao Data’s Gérard Thibault speaks to Louise Frampton about how the company is becoming ‘OCP ready’ OCP Data Centre Facility As part of this initiative, the OCP Data Centre Facility (DCF) project was developed to maximise the mechanical performance, and thermal and electrical efficiency of Open Compute Project servers. Research, conducted by IHS Markit and commissioned by OCP, has revealed that the main drivers of adoption include power efficiency and reduced cost, together with standardisation and quick deployment capability. Colocation spaces that can meet the demands of OCP and help cascade these benefits will

become increasingly important in the future. The DCF project focuses on five functional areas of the data centre: power; cooling; IT space layout and design; facility management and control; and facility operations. Forwardthinking data centre operators have the opportunity to become ‘OCP ready’ and among the first of these, in Europe, is Kao Data. The OCP Foundation is working with two data centres – one US and one European (Kao Data) – as part a pilot programme. Having joined the DCF Project, the recently completed London One facility has become the first

data centre to self-audit against Open Compute’s checklist for compliance with its engineering principles. The process will be complete when Kao Data presents its results during the OCP Regional Summit in Amsterdam on 1-2 October. Design principles Kao Data’s London One facility was designed and constructed to deliver power efficiency and resilience, low cost of operations (as a result of reduced electrical and mechanical engineering complexity), standardisation and fast deployment capability. After extensive onsite missioncriticalpower.uk

17 Having joined the OCP Data Centre Facility Project, the recently completed London One facility has become the first data centre to self-audit against Open Compute’s checklist for compliance with its engineering principles

Kao’s London One data centre facility is becoming ‘Open Compute Project ready’ assessments of the data centre, the data centre has been found to be structurally, technically and operationally capable of providing organisations deploying Open Compute Project designed solutions and equipment with an infrastructure platform that meets the central design concepts laid out in the OCP DCF project. “Hyperscale data centres have gained benefit from adopting OCP technology and, in Kao Data’s discussions with potential customers, OCP is increasingly a topic of conversation as, although initially an equipment missioncriticalpower.uk

driven concept, it has very real physical requirements from the data centre building itself. They can capitalise on the benefits driven into the servers and cabinets, by people such as Facebook, which means their IT platforms are more cost effective, and we can provide a data centre in which these OCP solutions sit comfortably,” comments Kao Data chief technology officer Gérard Thibault. He points out that many colocation data centres are not designed to cope with OCPtype IT infrastructure, including the weight of the cabinets (as many are delivered pre-built to

offer further efficiencies). “OCP racks commonly weigh around 1,300kg per rack. Locating and moving these on a raised floor can be logistically difficult. In addition, with weights moving towards 1,500kg, even details such as goods lift capacity and the gradient of ramps are important. The height of the OCP racks can also be a problem as these can go up to 58U. At London One, the facility has been designed to ensure that racks of this size can be received into the data centre. There is no raised floor (as it is constructed from solid concrete), so weight is not an

issue,” he continues. London One also meets the OCP objectives around efficiency. The facility utilises indirect evaporative cooling with no mechanical refrigeration to meet Ashrae TC 9.9 environmental standards and delivers an ultra-low PUE <1.2 even at partial load. Thibault points out that there are other energy efficiency benefits associated with OCP design. “With an OPC rack, access to the servers is from the front; as you do not need to access the back, you can drive up the hot aisle temperatures, which is much more efficient. OCP is about driving cost efficiencies – not just in terms of capex for the servers and racks, but also in terms of opex; they are trying to get people to realise that servers are quite happy operating in an environment that may stray up to 30oC, for a couple of hours per day – when it is warm. Historically, data centres tried to keep the data hall like a fridge and this costs money.” Kao Data also claims to go beyond the OCP requirements, in relation to the operating environment: “One key aspect, in relation to the warranty of servers, is the condition of the atmosphere – whether it has the potential to corrode silver solder and copper components on a circuit board, and lead to failures. As well as monitoring temperature and humidity, we monitor gaseous contamination, to ensure that warranties are not compromised, and that the environment stays within an acceptable range,” Thibault explains. Power is another key focus area for the OCP Foundation. London One is configured to provide up to 20kW per rack, with the infrastructure to match that capability, providing: • Availability up to 99.999% » August 2018 MCP

DATA CENTRE OPTIMISATION

• Back-up N+1 diesel generators, dedicated tanks ensuring operation at full load for at least 24 hours and a maximum of 48 hrs, with priority refuelling contracts in place • UPS – two independent (redundant) power supply systems available for a minimum of five minutes (at power failure) Thibault acknowledges that the preference for OCP is not to provide a centralised UPS. However, he favours a more pragmatic approach. “We are providing a flexible solution. We have a centralised UPS that can be used in the space, although OCP server racks offer the ability to ‘push’ the UPS element of the power system down to the rack level. Providing individual UPS is much more expensive than providing it on a room basis, in my view. But we can offer the option of providing generator backup power to the room, so that the OCP servers can use their local battery racks, or we can provide a central UPS facility if preferred,” he explains.

The London One facility adheres to OCP design principles around efficiency and resilience

The values of openness and collaboration, espoused by the OCP Foundation, will provide a ‘spur to innovation in the data centre sector’. The plan is for OCP to introduce a ‘stamp of approval’ for facilities – just as the Uptime Institute certifies data centres according to its Tier system MCP August 2018

Another key focus area of the OCP DCF project is facility management and control, which is primarily concerned with data centre information management (DCIM), building management systems (BMS) and power monitoring. “Kao Data has standardised on the Schneider Electric EcoStruxure Platform, which feeds back all the information on mechanical plant to our control room and raises alarms if there are any issues. There is also an opportunity to identify trends, so that we can fine-tune the operation of the data centre, particularly on the cooling side, and ultimately improve efficiency. On the power monitoring side, we use the Schneider PME software on the EcoStruxure system, and this helps us to identify any alarms and monitor the status of the electrical systems. It also gives us some trending information, such as the ageing of circuit breakers. This allows us to look at predictive maintenance to ensure we are always available,” comments Thibault. The final focus OCP area relates to facility operations, and Kao Data has been able to comply with OCP principles by using tools such as EcoStruxure with a team of engineers capable of interpreting and reacting appropriately to the data provided. Ultimately, Kao Data CEO Jan Daan Luycks believes that the values of openness and collaboration, espoused by the OCP Foundation, will provide a “spur to innovation in the data centre sector”. The plan is for OCP to introduce a ‘stamp of approval’ for facilities – just as the Uptime Institute certifies data centres according to its Tier system. Compliant facilities will be certified as ‘OCP ready’. The success of the project, in the long term will depend on the sector’s ability to embrace the values of openness and collaboration, and this will require a cultural shift. If colocation data centres take inspiration from the hyperscalers, and cascade this across the sector, the benefits could be significant. l missioncriticalpower.uk

VIEWPOINT

herever you look you will find predictions about ‘edge’ data centres – from the serious to the downright silly. The concept of edge computing is simple: the data traffic coming from the ‘Internet of Everything’ will overwhelm the communication networks, and putting the compute (and short-term storage) closer to the load will take the local strain, cut down the latency (delay in communication) and filter out the valuable data to be transmitted onwards and upwards to huge hyperscale data centres. These data centres will carry out all the ‘big data’ things – including (if you are cynical) selling your preferences, prejudices, moral weaknesses, movements and spending habits etc to anyone who wants to know about them in order to dislodge cash from your pockets. There will probably be a middle layer, operated by our present collocation data centres, where the connection from all the small, numerous and local ‘edge’ access points can be consolidated into a converged network of fat-fibre. The ‘Internet of Everything’ means exactly that: smartbuildings, transport, grids etc, personal video entertainment, numerous ‘phone’ apps accessing the internet and smartwear such as watches and glasses – too many to mention, half of which we can’t predict yet. The perfect edge application is the driverless car, clearly with lots of on-board computation capacity but also needing a wider vision of its world – not least traffic data for routing and road hazards to avoid, all in a split second. Would you feel comfortable as a passenger knowing that your safety depended upon a data connection to another country, or inside the Arctic Circle? Another application has already started to follow the road to edge: YouTube dominates the internet traffic MCP August 2018

On the edge of a new era? Ian Bitterlin makes some predictions on what the ‘edge’ will mean for jobs, energy consumption and where the ‘edge’ will be located

of the world and viral-videos could choke the network. As no one wants buffering for longer than a second (or nanosecond for some) so Google already pushes the most popular content to rented collocation facilities in the metropolitan areas to limit the long-haul from its remote behemoths. Yet it is simpler than that – we are not far from streaming UHD movies on our phones while travelling on the Underground. How could that be possible without a local source doing the buffering in the background for you with lovely advertising breaks? How many ‘edges’ could there be? The answer probably lies in a technology roll-out we have already experienced: for mobile phones why don’t we have one huge aerial in the middle of the country

(Banbury?) to which all 80 million SIM cards connect? Or for fixed line phones why not one huge telephone exchange? No, we have more than 40,000 mobile phone masts, each serving a small area (remember they were called cells?) and densest where the users are. Edge will be the same – densest deployments where the greatest number of applications (us) are. In a city like London, probably every 500-750m. How ‘big’ should they be? The ‘smart’ answer could be ‘40% smaller year-on-year if the load doesn’t grow’. In other words, the effects of Moore’s Law (in all its forms and derivatives) shrink everything over time. The only reason that data centres have grown is that demand has continually overhauled capacity – consider

the 2MW (2,000 x 1kW cabinets) ‘internet data centre’ built in the year 2000 that can now be outperformed by one 5kW cabinet. That is Moore’s Law in action. Anyone who predicts that edge facilities will be ‘X’ kW or ‘Y’ cabinets or ‘Z’ servers is very likely going to be wrong. I don’t mind being proved wrong, so I would guess that 20kW of the best hardware should suffice at the beginning and then refresh the components every 18 months. When I say best, I really mean fastest, since the difference between the best and the worst is a factor of 150:1 (read my last article in this journal) so that ‘X’ kW can mean everything, or nothing. The other point about size, or kW capacity etc, is that the edge facilities will serve very finite missioncriticalpower.uk

Would you feel comfortable as a passenger knowing that your safety depended upon a data connection to another country, or inside the Arctic Circle? areas with a certain population density, so that demand goes up it could be served by another edge installation, not just a fatter ‘edge’. So why 20kW? Well, we come to what I believe will be missioncriticalpower.uk

a key feature of ‘edge’. We are all concerned about the energy that data centres consume and so the thought of having 80,000 20kW facilities added to the power network (1.6GW at a PUE of 1.2 = 2GW = 5% of the UK grid) is not without concerns. The obvious advantage is to reuse the waste heat and make the edge facilities gridneutral. All we need is a nearby 24/7/365 heat load. Clearly the solution is to liquid-cool the 20kW of ICT load and take off the heat as hot-water at 65-70°C. House

the ‘edge’ in a cheap hotel room (which already will have a 5G aerial on the roof and a fibre connection), hospital, carehome, leisure centre or even apartment block and they can absorb 20kW (and a lot more) of hot-water for domestic services, 24/7. The host gets the rent and the hot water while the edge operator gets the ICT revenue, pays for the electricity and maybe even charges for the hotwater. The internet becomes carbon neutral, even ‘green’ for the first time. It is little different to city-centre buildings that already rent their roofs to the mobile phone operators. Just like mobile phone cells, if one goes down the users experience almost no difference as the nearest in the cellular network will cover the area, albeit with a weaker signal. But ‘edge’ will be smaller cells and so

fully skilled technicians in each subject area, power, cooling, ICT etc can guide the on-site technician ‘live’ on camera and head-up display. It does sound rather 1984, doesn’t it? But to us oldies so does the idea of a phone that is used for everything other than making phone calls – does anyone else remember how advanced we thought the ‘bleeper’ was? And then came along the bleeper with a singleline 16-character message display… Finally, we could not finish any discussion about anything ICT related, especially ‘edge’ facilities, without mentioning artifical intelligence (AI). Again, simple. We are 50-plus years away from AI (as defined by Alan Turing) as everything we now call AI is built on algorithms written by people and compiled on machines that use software written by people.

redundancy in service coverage will be more that sufficient. With ICT loads already capable of remote diagnostics, upgrades and re-boots (as most collocation users already can) the edge facilities will be dark and unmanned. We will create 60,000 new jobs (20,000 on three-shifts) of mobile technicians. In view of the skill shortage we currently face, these mobile techies could become semiskilled – wearing cameras with a headset and 24/7 access to a technical help centre (hopefully not overseas) where

The junior brother of AI, machine learning, is also based on code written by people… People are not artificial, yet, but the key word is ‘people’ and, at the end of the day, that it is one resource that we have too much of and the last thing the economy of any country wants is unemployed people. Make no mistake, ‘edge’ is logical, sensible, required and unavoidable and gives us a great opportunity to get the reuse of waste ICT heat, at last, possible for cities that don’t have district heating – that is, most cities. l August 2018 MCP

UNINTERRUPTIBLE POWER SUPPLIES

Power reliability under the microscope Research conducted using a giant microscope could provide a solution to cleaning up the world’s plastic pollution problem, as well as having an important role in tackling serious threats to human health. Interruptions to power have the potential to damage sensitive equipment and disrupt this vital research

ocated at the Harwell Science and Innovation Campus in Oxfordshire, Diamond Light Source is the UK’s national synchrotron facility. The not-for-profit company, a joint venture funded by the UK government through the Science & Technology Facilities Council (STFC) in partnership with the Wellcome Trust, is used by more than 7,000 researchers from both academia and industry to conduct essential experiments that keep the UK at the forefront MCP August 2018

of scientific research. At the heart of the worldleading research centre lies a giant machine, half a kilometre in circumference, called a synchrotron. It works like a massive microscope, harnessing the power of electrons by sending them through three particle accelerators to near light speeds so that they give off light 10 billion times brighter than the sun. These intense beams, predominantly in the X-ray region, are then directed

into laboratories known as ‘beamlines’, of which there are currently 31 operational on site. Scientists use them to study a vast range of subject matter down to the level of atoms and molecules. The applications are broad, reaching from smarter materials, to better data storage, to clever clean-up solutions for plastic pollution, as well as the preservation of significant historical artefacts such as the Mary Rose. Synchrotron light has also played a pivotal role

in the study of diseases such as foot and mouth and polio, vital antibiotic research and other areas of medical science. Downtime for synchrotron The beamlines operate at the micron level and below, and are incredibly precise. To achieve this precision, each one has to be calibrated through several items of equipment. If power were to be lost, or even slightly interrupted, they need to be completely recalibrated. This would take between one and two weeks for each beamline, losing valuable research time. Diamond Light Source has also recently become the UK’s largest electron microscope hub, with eight top-of-therange microscopes on site. The heating element in the tip of each microscope is a sensitive filament. If the power is turned off without letting it cool down, it will be destroyed and need replacing. Because of the importance of the research being undertaken missioncriticalpower.uk

Diamond Light Source at Harwell Science and Innovation Campus is the UK’s national synchrotron facility

The heating element in the tip of each microscope is a sensitive filament. If the power is turned off without letting it cool down, it will be destroyed and need replacing and the sensitivity of the equipment, the facility requires robust and reliable critical power protection at all times. Diamond Light Source uses an uninterruptible power supply system to ensure continuity of supply. This UPS system needs to be reliable, with dependable ongoing maintenance support to quickly alleviate any future faults or failures. Riello UPS had the solution. The challenge The project needed to be completed quickly and efficiently over a six-month period, working to a timeline of eight weeks on and three weeks off to ensure minimum downtime. As a scientific facility where about 600 staff are located, there was a challenge in the amount of input needed from different sources. Andy Reed, head of electrical facilities at Diamond Light Source, explains: “Due to the nature of the facility, we have limited downtime to play missioncriticalpower.uk

Specialist Power Systems was commissioned to design and install a bespoke UPS to ensure ongoing critical power protection across the complex site. With beamline time at a premium, ongoing power availability is crucial in supporting the integrity of Diamond Light Source and the services it provides.

A Multi Sentry solution Following an extensive review of the facility’s requirements and input from those who work there day-in, day-out, Riello UPS’s Multi Sentry was selected as the most suitable system. Lee Scott, senior sales manager at Specialist Power with. With eight weeks of solid Systems, comments: “After running time followed by three visiting Diamond Light week-long shut downs, we knew Source and discussing their it would be a real challenge to requirements, Riello UPS’s Multi replace the existing systems to Sentry was the obvious solution. the standard required in the It provides high performance time provided. and efficiency levels, within a “The existing UPS was a very compact footprint and has ‘closed protocol’ system, meaning the flexibility and the units were reliability required.” ‘locked out’ The state-ofto anyone the-art UPS has a other than the high level of online manufacturer. double-conversion This had caused performance. It is several issues specifically designed since the initial to protect critical installation many information and years ago, so it telecommunications was important systems, networks, that we weren’t services and limited to one processes where maintenance operations could be option going disrupted by poor forward and power quality or the new system breaks in the mains from Riello power supply. offered a more In addition to open approach.” reliability and value Certified Riello Riello UPS UPS reseller Multi Sentry for money, the Riello

UPS Multi Sentry offers an advanced communication ‘open protocol’, making it easier for those working onsite to monitor loading and battery autonomy. The large front panel graphic display provides vital information, such as real-time measurements, operating status, and alarm conditions. The UPS installation was completed three months earlier than the expected six months, which meant significant cost savings and reduced downtime for the facility. Another added benefit was the fact that no power load was lost during the implementation. Since the initial implementation, the facility now operates 41 Riello UPS units, 13x 120 KVA units and 28x 30 KVA units. Diamond Light Source uses a parallel UPS design for complete protection against power failure. For example, if two UPS units are needed to support the critical load, they will run three in the unlikely scenario that one fails. Reed concludes: “In recent years, UPS failure has been a real issue for us and has had a knock-on effect with the laboratories and scientists we work with. However, with the new Riello UPS installation, we have had total confidence from start to finish and also feel we have an added level of protection with the ongoing maintenance support. “We use Riello UPS’s PowerNetGuard monitoring software to enable us to view the status of the units at all times, and Specialist Power Systems also provide continued onsite maintenance which is undisruptive to our work.” l August 2018 MCP

THERMAL MANAGEMENT

s demand for data cost and energy savings if the collection, storage and right combination of server exchange multiplies at technology and cooling plant are an unprecedented rate, experts specified. are concerned that the energy The reality is, thanks to consumption of data centre advances in the operating infrastructure is becoming an parameters of server increasing drain on the world’s technology, working with energy resources. cooling experts early on in the In an era where, on an annual design phase of a new data basis, data centres in the US centre campus or hall, can alone consume the entire deliver considerable capex output of 34 of the savings by completely world’s largest eliminating the power stations, need to purchase methods of traditional conserving refrigerated and reducing cooling systems. the energy In other words, input of some up until now, of the globe’s specifiers would most socially and select an air economically critical conditioning system buildings is more which would require Tim Bound important than ever. additional cooling plant By 2020, the UK will be (typically an ambient cooler) to the largest single market in dissipate the heat generated by Europe for data centres, and the compressor. Now, you may analysts predict this will cost only need the ambient cooler. the industry up to £7bn per year in energy alone. Disclosure of Why is ambient cooling data centre energy and carbon increasing in popularity? performance metrics is now Historically, data centres have driving change in a sector where been cooled using compressor the potential energy savings or refrigerated technology, could comfortably sit in the often with an adiabatic hundreds of millions. cooler installed to dissipate There are a number of ways the heat generated by the to deliver power savings in compressor. The performance data centres, from optimising of this conventional chilling renewable energy sources, technology has serviced the to updating the physical mission critical data centre infrastructure of the building, sector well, thus far, but change investing in modern server is definitely afoot. technology with lower There is a distinct shift in heat emissions, or carefully popularity from traditional specifying the right cooling compressor based chilling equipment. methods to far more energy When it comes to designing efficient ambient cooling a new data centre campus technologies, with a particular or delivering the latest build focus on adiabatic solutions due out phase of an existing asset, to its retrofitting capabilities there are opportunities for vast compared to other ambient

As the data centre industry evolves and grows at a rapid pace, energy conservation focused industry leaders like Facebook have already maximised ambient and free cooling technologies within their data centres MCP August 2018

Is ambient cooling overtaking compressor technology? Tim Bound, director at Transtherm Cooling Industries, explores the evolving role of ambient cooling technologies in keeping energy costs low in data centre applications

systems such as direct evaporative or heat exchanger technology. Why? The two main drivers for change are a surge in investment in the construction of new or extended data centres, plus a significant improvement in the operating parameters of server technology. The global data centre construction market is estimated to grow from $14.59bn in 2014 to $22.73bn by 2019, at a Compound Annual Growth Rate (CAGR) of 9.3%. This thriving market growth is down to many companies transforming traditional facilities into mega data centres and others planning to build new ‘monster sites’ in the coming years. Server technology With modernisation, comes

state-of-the-art server technology, which is capable of withstanding higher temperatures while maintaining optimised performance and reliability. The modernisation of legacy data centre infrastructure, the increased construction of new ultra-modern sites and the widespread adoption of more temperature resilient servers has driven a desire for compressorless cooling, choosing the energy efficient benefits of ambient cooling technologies instead. Why is this happening now? Historically, ambient cooling equipment has been unable to cool data centres to within the right temperature range, which missioncriticalpower.uk

ÂŁ7bn

The predicted cost of energy for UK data centres by 2020. Potential energy savings could sit in the hundreds of millions

created an industry reliant on refrigerated or compressor-led solutions. Now, thanks to more adaptive build methods and the marginal, but vital, increase in server temperature resilience, the temperature parameters of data centres has risen to a range which ambient cooling technology can comfortably achieve. In other words, instead of cooling the compressor, adiabatic equipment is now directly cooling the data centre equipment itself. This is reducing capex expenditure by eradicating an entire chiller from many new build specifications. Why is this a more energy efficient solution? Adiabatic and other ambient cooling equipment achieves missioncriticalpower.uk

lower energy efficiency ratios (EER) than compressor-based chillers. Take the example of a typical compressor chiller which will consume 1kW of energy for every 3-4kW of cooling delivered. Compare this to an equivalent sized adiabatic solution, which for the same 1kW of energy consumed, will deliver up to 75kW of cooling. In terms of EER, this difference translates as an approximate EER of 4 (4kWth/1kWe) for a conventional compressor solution, and an impressive EER of around 75 (75kWh/1kWe) for the equivalent ambient solution. This, of course, translates to substantial energy and cost savings for energy-hungry mission critical environments.

Can legacy data centres still benefit from ambient cooling? Older data centres that house older style servers will still rely on compressor or refrigeration chillers to dissipate enough heat from the data centre, and maintain an internal building environment which meets the operational temperature thresholds of dated server technology. That said, given that most leading ambient coolers are installed externally, retrofitting additional cooling technology to supplement an existing compressor solution is a highly viable option for many older data centres looking to maximise geographical assets and weather patterns to deliver

free cooling during certain seasons. If an older data centre retains its legacy infrastructure but is refitted with state-of-the art servers which can operate at higher temperatures, adiabatic cooling equipment can be retrofitted to the same inlets as older equipment. As the data centre industry evolves and grows at a rapid pace, energy conservation focused industry leaders like Facebook have already maximised ambient and free cooling technologies within their data centres. As we move towards a future even more reliant on the storage and exchange of data, the rest of the industry is expected to follow suit. l August 2018 MCP

POWER DISTRIBUTION

Powerful intelligence: the role of smart PDUs The increased sophistication of PDUs is in line with expectations for increasingly granular data. Panduit’s Sander Kaempfer provides an insight into the role of intelligent technologies in the data centre and the scope for smart PDUs

ower demands at data centres continue to grow, with the number of global sites increasing steadily and medium-sized sites requiring up to 8MW, guaranteed supply from the grid and significant investment in redundant primary switchgear, together with the assortment of power generators and UPS to ensure the IT environment provides continuous uptime. Those are the big numbers but what about at the cabinet level power distribution? Innovations in power distribution units, what some are calling intelligent PDUs, are creating live systems that are generating actionable data. ‘Take care of the pennies and the pounds take care of themselves,’ is a highly relatable idiom for anyone designing power distribution infrastructure for a data centre. One of the largest operational expenses for a data centre is energy. It is estimated that on average 27% of ongoing costs MCP August 2018

are on power, and operators and customers want as much of that energy as possible to be used by productive IT systems. How do we do this when IT load has become less predictable, with massive change in usage driven by the growth of mobile data, IoT and media? Factors driving interest in intelligent or smart PDUs include the increasing demand for high-power capacities in high-density compute environments and rising demand for intelligent infrastructure products in data centres. This increase in server density in the racks is key to the intelligent PDU market, which was valued at $1.20bn (£910m) in 2016 and is expected to reach $2bn (£760m) by 2023, illustrating the industry’s drive to acquire systems that can positively integrate into managed IT environments. In the same period, the high density rack server market will grow from $36.5bn (£27.7bn) in 2016 to $90.56bn (£68.8bn).

PUE metric: tick box or key achievement? The Green Grid developed the power usage effectiveness (PUE) metric more than 10 years ago. This is a ratio of total annualised facility energy use against total IT energy use, and provides the market with a comparative measure of the data centres energy efficiency. The industry’s average PUE for a medium-sized facility is 1.7, and there is momentum within the industry to design and build sites with PUE of about 1.2-1.4 (for example, Kao Data London One data centre site has a PUE of <1.2). A key route to reducing energy use in the compute area is to accurately measure, analyse and control power to the heart of the white space. Taking an increasingly granular view of IT power use through rack-mounted intelligent PDUs and their management systems allows operators and customers to control the IT environment in a way not previously available. Even with the growth in high-

Intelligent PDUs such as Panduit’s G5IPDU automatically gather and analyse data and can control individual outlets. The intelligent PDU market is expected to be worth $2bn by 2023

Figure 1: Intelligent PDU and rack server market values

Intelligent PDU

$2bn

$1.2bn

2016

2023

missioncriticalpower.uk

27 density servers and the additional power draw on a per rack basis, the control now available can substantially lower the overall power requirement of the customerâ&#x20AC;&#x2122;s computing load. The need for remote power monitoring and management has been driven by the change in strategy in enterprise computing, and the growing use of colocation by the enterprise market, and has now become an essential requirement for many organisations, as part of Cloud or hybrid IT strategies. The latest generation intelligent PDUs offer sophisticated features at the rack-level, which can automatically gather and analyse granular data by monitoring software. Energy metering and monitoring provides watt-hour energy billing capabilities, accurate enough for customer billing, while outlet metering offers information required

Rack server $90.5bn

$36.5bn

2016

2023 Source: Panduit EMEA

missioncriticalpower.uk

for Green Grid Level 3 PUE calculations. Input-metered PDUs determine the available capacity of the racks, and outlet-metering determines the available capacity to facilitate system provisioning. User-defined criteria allow input-metered PDUs to trigger alarms, providing warnings about potential circuit overload, therefore eliminating most IT failure scenarios, while outletmetered PDUs are used to measure the actual consumption of power at the device and server levels, a requirement in colocation facilities where multiple customers can share cabinet space.

servers to be operated up to their manufacturerâ&#x20AC;&#x2122;s optimised performance temperatures. These highly reliable power systems can monitor and detect power fluctuations and implement phased reductions and shutdowns on individual or groups of servers, and raise alarms and notifications if manual intervention is required. The increased sophistication of the PDU is in line with expectations for increasingly Measurement and control granular data that is being Rack PDUs that have the ability monitored and analysed around for outlet level power monitoring the system. and outlet level switching Standard PDUs are not capabilities for individual outlets designed to integrate with data or a group of outlets have the centre information management capability to identify idle servers systems. Hyperscale level that are still consuming power environments, or greenfield data without contributing to the centres being developed to those IT load, these are often called levels of sophistication, need the zombie servers. This capability capacity to interrogate compute provides management with the and environmental systems at data to consolidate applications the rack, in order to optimise across fewer servers, reducing control of the total system. overall power use and increasing This is an important element cost efficiency. in the challenge to be flexible Power monitoring at to customer requirements, the individual outlet level while maintaining a policy of provides increased amounts of increased energy efficiency and actionable management data site productivity. As the cost on the power consumption of data centres increases and of each connected IT device the environment become more enabling the redeployment or complex, with taller cabinets, decommissioning of individual increased numbers of servers, servers, storage or switches to operating at high temperatures, reclaim under utilised power intelligent PDUs will become capacity. increasingly integrated with Smart PDUs offer the ability more of the smart infrastructure. to control individual outlets, Control of energy use is the allowing remote rebooting of largest potential cost saving IT equipment if necessary, and available to the data centre power sequencing to ensure operator. The increasingly power fluctuation and surges sophisticated smart technologies are eliminated during shutdown being incorporated into building or power up. This will improve infrastructure are designed problem resolution response to access massive amounts of time and, in many cases, reduce data, analyse it in real-time the requirement for physical and take actions to mitigate technical support to attend an potential incidents. This incident. growth in operational New developments data analysis ensures have created reduce overall cost PDUs with high reductions and temperature positively effects grade the continuous components reduction to operate up in the PUE to 60oC for measurement, extended periods, while maximising which work within individual customer the latest Ashrae 9.9 guidelines allowing Sander Kaempfer information.l

Control of energy use is the largest potential cost saving available to the data centre operator

August 2018 MCP

BACK-UP POWER

The ‘unsung heroes’ you can bank on How do you know your back-up power will kick in when it is needed? A recent report suggests there is unrealistic optimism about resilience. Paul Brickman of Crestchic examines why critical sites should consider investment in loadbanks as part of their strategy to ensure uptime

ith recent research says that businesses are risking from Centrica stating £2.8m of annual revenue through that 33% of energy lack of energy resilience, so why decision-makers believe their is there such a high percentage organisation is not prepared of businesses with no plan in for a disruption to their energy place? People often think it will supply from a temporary grid not happen to them. It may come failure, there are calls for as a surprise to many but more attention to be it is much more likely paid to the role that an outage will loadbanks play in occur now than 20 energy resilience years ago. This is strategies. because there is The less of a buffer of Centrica report power available predominantly nowadays. highlighted the As highlighted disparity between by National Grid’s the threat posed recent report on Paul Brickman by energy-related frequency response failures and the priority that reserve, the amount of system businesses give to having a inertia has been reducing in strategy for a response in place. recent years, which results in Loadbanks are a simple and the system frequency becoming effective solution. more volatile closer to real-time While most businesses and, therefore, susceptible to understand the importance potential brownouts. As the of having a robust back-up Centrica report states, this generator in place, they do not is due to the replacement of consider what would happen steady ‘base-load’ generation if the back-up generator didn’t methods with more sustainable, meet the critical loads it is renewable energy sources. supposed to, when called upon. National Grid’s report outlines That is where loadbanks come its existing Enhanced Frequency in – they are the unsung heroes Response (EFR) product will of energy back-up. improve operation of the system, Loadbanks develop an but in its current form is not electrical load which is then necessarily the optimum design applied to an electrical power of a fast-acting product. While source that can effectively decreased system inertia is simulate power failure under full high on the agenda of National load demand, thereby ensuring Grid, there is not yet a robust that all generators, controls solution in place, making it even and switch panels function as more crucial that businesses required in real conditions. implement further measures to There is certainly unrealistic protect their power systems. optimism when it comes to Given that the cost of a energy resilience. The headline loadbank is typically only a finding from the Centrica report fraction of that of the systems MCP August 2018

which it supports, it makes sound commercial sense. There are some questions that we are frequently asked about loadbanks and this article will seek to address some of these: Is it sufficient to just have a generator/UPS for backup power for mission critical facilities? A loadbank should always be considered if you wish to safeguard the reliability of standby power systems. If you have invested in any form of uninterruptible power supply and/or diesel generator, then

Loadbanks in situ missioncriticalpower.uk

29 needs to consider the use of a resistive-reactive loadbank. What other benefits can a loadbank offer? Battery optimisation is one of the biggest benefits that loadbanks can offer. The loadbank can enable a storage system to participate in both low and high frequency demands, while ensuring the battery remains fully charged. This helps significantly reduce the capital expenditure on the main battery itself. By matching a loadbank with a battery storage device, it allows the battery to remain close to 100% charged and therefore increase frequency response revenue by accepting the full income against the battery’s rated output. In times of high frequency demand, the battery simply accepts the charge for

operator to state the emissions output more often than they have done in the past. There is an exemption for plants that are used for backup generation, provided they are operated for no more than 50 hours per year. This will be lost if they enter into a capacity or balancing service agreements. The use of a loadbank can provide a load to simulate the conditions the generator will experience when exporting to the grid, or when providing a commercial or industrial site with all its power in island mode. What are some of the most common myths and misunderstandings around loadbanks? One of the most frequent is that loadbanks cannot operate in the rain. The construction of a Crestchic

Renting a loadbank is a relatively modest investment that could help save your business in the long run from disruption and a complete shutdown you will find that the cost of a loadbank is considerably lower. It is imperative you use the loadbank to test those systems to make sure they work. Any generator that spends its life ‘on standby’ is a potential timebomb, if not regularly tested under load. Simply running the generator to idle a few times a month is detrimental to its ability to perform when called upon. In fact, you’ve just risked your investment in the back-up power, risked your entire business operation and potentially risked lives. This is particularly the case for offshore emergency systems, hospitals power supplies or fire suppression systems that are driven by the emergency power facility. How can the public sector justify a business case for investment in loadbanks? There is a real argument here: you don’t have to invest in loadbanks. Crestchic rents loadbanks, as well as selling them, which provides a certain missioncriticalpower.uk

amount of flexibility for businesses. A loadbank is worth the money invested into it. If you do not test your emergency power system, or if you were to test it in a poor manner (ie no load) you are reducing the value of the asset much quicker. If you look at the bigger picture, renting a loadbank is a relatively modest investment that could help save your business in the long run from disruption and a complete shutdown. A resistive loadbank is fine for testing the engine but when there is a need to test the alternator by the introduction of a lagging power factor, then the operator

33%

of energy decision-makers believe their organisation is not prepared for a disruption to their energy supply

an initial 8-12 seconds until the loadbank takes over indefinitely draining the battery of its charge. How will recent legislation around emissions and generators influence the load bank market? The Medium Combustion Plant Directive (MCPD) will see an increase in the need to test generators under load. The directive imposes limits on emissions from mid-sized generators. These emissions include sulphur dioxide, nitrogen oxides and dust from plants with a thermal input of at least 1MW and up to 50MW. For newer plants, the limits will come into effect from 20 December 2018, and for existing plants the limits will apply from 1 January 2025. However, this limit will only apply if they have thermal input of more than 5MW, and from 1 January 2030 if they have thermal input of up to 5MW. This means there is an increased responsibility on the

loadbank ensures that it is completely weatherproof. Many of the larger units are shipped around the world via shipping containers, so they are open to all elements such as changes in temperature and loading and unloading from ship to carrier. We are also regularly surprised at the amount of people who think they must apply the load manually. They are even more surprised to learn the loadbank can be pre-programmed with various load profiles over a given timescale, leaving them free to monitor the generators’ many parameters, such as temperature, exhaust gases and vibration. Voltage is also misunderstood. A loadbank must be specified to the voltage output of the power source. While the loadbank can accept lower voltage, it won’t take kindly to someone plugging in a much higher voltage. You wouldn’t plug your TV into a 415V three-phase supply, or the 33,000V overhead cables and expect it to work, would you? l August 2018 MCP

POWER MANAGEMENT

Healthy returns: delivering energy savings and resilience An upgrade of backup power systems is enabling a major hospital in the US to reduce its energy costs. Could UK hospitals follow suit?

ncreasing reliability, redundancy and flexibility were the key goals for Rex Hospital when it decided to upgrade the backup power systems for its not-for-profit healthcare facility. Based at Raleigh, North Carolina, in the US, the hospital treats tens of thousands of inpatients every year. As its facility continued to expand, Rex looked to upgrade its existing open transition power system design, which included an interruption of service during the transition between utility power and generator power. The system also relied on generators and fuel tanks on flatbed trucks to provide additional capacity during construction or when adequate power could not be delivered to the hospital load. Facility services director Mike Raynor proposed a fail-safe, closed transition system that would allow for a transfer between utility and generator sources without interruption of power to the hospital. The team understood the advantages of the closed transition design and convinced management that the slightly higher first cost of a closed transition system would deliver cost savings over the life of the system and would be worth the investment over the long term. They successfully presented their case to the hospital’s executives, medical staff and regulatory officials. MCP August 2018

The design implemented meant replacing the utility substation and making it more reliable, as well as relocating the switches and switchgear from cramped quarters in the main hospital building to a newly constructed central energy plant. The entire project and system switchover was completed with only a single, planned 10-second outage. The plan takes anticipated growth into account, with enough emergency capacity (8.25MW) to handle a proposed seven-storey heart centre and future cancer centre addition. Rex uses an N+1 arrangement – which means it can take one generator out of service and still retain adequate capacity. The plan replaced three 1.25MW generators with two Caterpillar 3MW generators, and kept an existing Caterpillar 2.25MW generator. There is room to add more switchgear and circuit breakers. An automatic transfer switch and an uninterruptible power system have been added to protect the hospital’s data centre. There are two 150,000-litre underground fuel tanks, and the system maintains fuel in each generator’s emergency 565litre ‘day tank’ at all times. Fuel capacity for the previous system was 227,000 litres – one-third less than the new system. With all tanks full, the hospital could meet its own peak demand

(about 5,200kW) for almost six days. However, since that peak is reached only for short periods on the warmest summer days, the hospital could probably operate under its own power for more than nine days for much of the year. The hospital’s new substation consists of four utility-owned, pad-mounted 2,500kVA paralleled transformers providing a total utility capacity of 10,000kVA (10mVA). The hospital assumes ownership at the transformer secondaries, which are connected to the hospital’s outdoor switchgear. When an outage occurs, the switchgear

automatically disconnects from the utility by opening four 1,200 amp circuit breakers, and simultaneously sends a signal to start the generators. Based on its present peak load, the hospital can continue to operate without interruption should there be a loss of one transformer. If two or more utility transformers were lost, the hospital’s generators will start and parallel while the outdoor switchgear disconnects from the utility system. The hospital will then remain on the generator source until the utility source is restored, at which time the generators will parallel with the recovered source. Once the utility voltage has stabilised, it missioncriticalpower.uk

The Scada system provides interactive monitoring, realtime and historical trending, distributed networking, alarm management and comprehensive reports around the clock for every detail of the entire power system

will reconnect to the hospital load without interruption. The utility’s transformer primaries are served by two 25kV utility feeders from separate distribution systems. Though both are energised, the hospital can draw from only one at a time. If the active feeder is lost, the utility can manually switch the hospital to the backup 25kV source at the hospital’s substation. Another important feature of Rex Hospital’s comprehensive power system is the Scada system, designed by Russelectric. The company’s systems can provide sophisticated control functions, including emergency/standby missioncriticalpower.uk

power, peak shaving, load curtailment, utility paralleling, cogeneration and prime power. The Scada system provides interactive monitoring, realtime and historical trending, distributed networking, alarm management and comprehensive reports around the clock for every detail of the entire power system, not only the backup components. With this system, technicians can fully monitor and control the entire power system from the control room at the central agency plant. An operator uses full-colour ‘point and click’ computer-screen displays at the system console to access and change the system’s PLC setpoints, display any of the analog or digital readouts on switchgear front panels, run a system test, or view the alarm history. In the event of an internal failure, the Scada system can rapidly and automatically configure a path to bypass the failure and re-energise the system without starting the generators. The Scada

system’s full manual backup was another key advantage. If the touchscreen fails, operating personnel can manually open and close breakers, synchronise and parallel the generators onto the bus, and add or shed load. Other manufacturers’ systems do not provide for full manual operation. The Scada system includes a simulator that shows trainees what to expect when they lose a feed, open or close a breaker, or add or remove load. The simulator uses the same control logic software as the switchgear’s programmable logic controllers. The crew also uses the simulator during startup and for trouble-shooting, system improvements, preview testing, and tours. Peak shaving The new system also enables the hospital to perform peak shaving, supplying some of the hospital’s power while the utility is supplying the rest, thereby saving on utility demand charges. The system does not contribute power to the grid but

its load curtailment capabilities means it can respond quickly if the utility asks the hospital to reduce demand on the grid by a specified amount. The resulting contractual rebates lower the hospital’s overall energy costs. For example, if the utility experiences an unusually high demand for power for air conditioning during a heat wave, under their contract it may ask the hospital to generate its own power for a specified amount of time. On average, this type of request happens only once or twice per year. The new system offers greater reliability, more redundancy and increased flexibility, as well as reducing energy costs for the hospital. “The hospital needed a new and modern system that built on what we had already,” comments Raynor. “Working closely with Russelectric, we came up with a very sophisticated system, and we’re at a point now where the system is functioning as we expected. All the hospital’s electrical needs are covered.” l

The new system enables the hospital to peform peak shaving, supplying some of the hospital’s power while the utility is supplying the rest, thereby saving on utility demand charges August 2018 MCP

DEMAND-SIDE RESPONSE

Making money from your UPS Vertiv’s Valerio Zerillo and Emiliano Cevenini explained ‘how to earn revenue from stored power in UPS systems’ as part of a recent webinar. The rewards, they claim, are compelling – even for large data centre operators. Louise Frampton reports

he integration of renewables is presenting increasing challenges for National Grid, according to Vertiv’s Valerio Zerillo, a product manager for the AC power division. He believes demand-side response – using UPS and generator assets – could help solve some of the issues, while reducing energy costs for consumers. Data centres and critical facilities, in particular, are ideal candidates for participation in balancing services, says Zerillo. Increasingly, there are two-way flows of power. Customers are evolving from being passive consumers of power to becoming active ‘prosumers’. Zerillo explains that demand can be turned up or down to achieve a balanced,

secure network and the UK is emerging as a strong near-term market for distributed [behindthe-meter] energy storage. Aggregators are allowing participation of small energy storage units, including UPS, in the electricity market and this can be performed using Vertiv’s platform. The company’s UPS and battery energy storage technologies enable power intensive users – such as data centres, telecom sites, and commercial and industrial facilities – to optimise their power assets to save on energy bills and earn additional revenue. The technologies also offer the opportunity to efficiently utilise solar and wind energy for local usage, storage or for export, helping to preserve the

energy availability and resilience of the site. Data centres can achieve additional revenues from grid services, while the UPS primary mission is untouched. Vertiv vice-president Emiliano Cevenini pointed out: “Vertiv’s primary objective is to protect the load. The UPS must provide critical backup power and quality functions, but the next generation of UPS can deliver other services, in addition to this primary mission – such as demand management, energy arbitrage and grid services. This can reduce the cost of energy for consumers while helping to ‘smooth out’ demand for the grid.” Citing figures from a reference data centre site based in Ireland (see table 1), he claimed that the revenues generated by stacking

services can be significant. Using diesel generators and UPS, the total revenue stream could be as much as ¤1.97m (£1.75m) per annum (based on market revenues for the start of 2018). “Even for a big data centre operator, these numbers are interesting,” says Cevenini. “Colocation data centre operators have a competitive business model where the price per square metre they can offer to the customer is key. These types of additional revenues become interesting providing we can ensure that the primary mission is met, so that service level agreements are maintained with their customers.” He adds that lithium-ion batteries are the preferred solution for battery energy storage applications, due to their: • Long levels of cycles • High efficiency • High energy density • Ability to fully discharge without affecting battery life • Quick charge and discharge • Lower price level compared to the past In conclusion, Zerillo and Cevenini argue that data centre operators have the opportunity to monetise an existing asset, while keeping the critical load protected. The latest technology helps data centre operators to participate in supporting the grid and this is becoming crucial as renewable energy sources are challenging traditional electricity production. l

To view the webinar in full, visit: https://tinyurl.com/yc4cbq3f

Reference data centre – Ireland

Data hall

UPS

20MW with five-minute autonomy in N-2 arrangement

Adressable load

9,600kW

Available UPS energy

320kWh

Contract size

9.6MW

Alternative contract structure (annual incomes on five-year contracts) 0.5-10 seconds

562

Primary Operating Reserve

5-15 seconds

495

15-90 seconds

299

90 seconds - 5 minutes

237

5-20 minutes

189

Indefinite

-39

20 minutes - 8 hours

154

1,395

1,975

Tertiary Operating Reserve Tertiary Operating Reserve (manual) Steady State Reactive Power (reactive) Diesel generator income

Source: Vertiv

MCP August 2018

UPS and diesel generators (¤000s)

Fast Frequency Response

Secondary Operating Reserve

Table 1: Application case – data centre

UPS only(¤000s)

Total site revenues per annum

missioncriticalpower.uk

DATA CENTRE DESIGN

Fast track to reliable power Angel Trains has chosen to operate its own on-premise data centre. Remote monitoring and diagnostics help to ensure high levels of resilience and efficiency

ngel Trains is one of by Angel Trains are its assetBritain’s leading train management system, a bespoke leasing companies, application developed in-house providing rolling stock to that manages the company’s several of the UK’s largest inventory of rolling stock, and train operating companies Oracle Business Suite, which including Virgin Trains, Abellio comprises the financial stack Greater Anglia, Arriva Rail of software including accounts North, Great Western Railway receivable, general ledger and and South Western Trains. invoice-management. Established in 1994 as part “Together those two of the process to privatise applications are the most British Rail, the company now business critical, being employs approximately 150 responsible for managing people split between its primary our revenue generation and headquarters in London’s collection, on which the business Victoria and secondary premises depends on,” says Wren. in Derby. “However, we also run several The company owns some other Microsoft server-based 4,000-plus rail vehicles, which applications such as Sharepoint it leases to operators on terms for our content and document that are generally coterminous management.” with franchises granted by the Department for Transport. Agility is key “We’re a big-ticket asset leasing The key priorities for the data centre, in which the applications company,” says Andy Wren, head of IT services. “We have an are hosted, are agility, reliability and cost-effectiveness. intricate business model and our IT systems that support it are Although the company tries to similarly complex.” standardise leasing contracts Angel Trains’ IT department with its customers, the reality is that each agreement has an comprises eight people, element of customisation, with including Wren, and is based consequent demands on the at the London HQ where the corporate data centre is located. IT department’s development Wren is responsible for effort. An implication for the leading the entire IT data centre is that it must have the agility services function to scale up capacity including to accommodate application additional development, software servers, should procurement, they become support for users necessary to meet customer and management requirements. of the data-centre For reliability, the data infrastructure that underpins it all. Andy Wren, head centre has a south The key IT London-based of IT services at disaster recovery site systems operated Angel Trains MCP August 2018

to which all its data is replicated and securely backed up. Driving all the IT investment decisions, however, is the perennial need to keep costs low while maintaining a consistently reliable level of service. Angel Trains’ IT department enjoys the convenience of being able to run its own systems on-premise, but IT management is cognisant of the fact that thirdparty service providers can offer hosting services from remote sites at competitive prices. The ownership, control and speed of connectivity from the on-premise solution has many benefits for the company, of which one is avoiding any latency issues, particularly with large files. Owning versus outsource “As an internal IT department, we are comfortable with having the ability to monitor our own missioncriticalpower.uk

35 We investigated a number of different hosting partners and found there is a great variety of services available. However, it was more cost-effective to own and manage our own on-premise data centre

infrastructure and IT equipment, rather than have a third-party managing it on our behalf at another location,” Wren says. “We investigated a number of different hosting partners, and found there is a great variety of services available. However, it was more cost-effective to own and manage our own on-premise data centre.” Among the service options considered were simple colocation, where the servers could be hosted in a colocation site operated by a third-party company but Angel Trains’ staff would continue to manage the systems remotely. Alternatively, some or all of the management of the systems could also be outsourced to an external service provider. However, the company decided to continue to operate its data centre in-house, with the help of a maintenance and missioncriticalpower.uk

support contract with APC by Schneider Electric Elite Partner, Comtec Power. Added resilience Angel Trains has been using Schneider Electric UPS systems for 10 years, attracted initially by what its UPS products offered in terms of flexibility, with the ability to perform ‘hot swaps’ of components such as batteries and power controllers. Its data centre comprises a rack-based containment system, with critical power protection provided by Schneider Electric Symmetra PX UPS units. For additional resilience, there is a dual power feed running direct from the mains and an emergency backup power generation unit on-site. With key challenges that included cost effectiveness, reliability and footprint, in terms of space,

Angel Trains chose to adopt Schneider Electric’s ISX Pod architecture with InRow cooling for its data centre. “Once we were introduced to Schneider Electric’s on-demand InfaStruxure solution with InRow cooling, we knew that was exactly the type of architecture we wanted to move forward with,” says Wren. “We needed to make the new data centre as cost-effective, scalable and robust as possible and the Schneider Electric racks and Symmetra UPS systems hit the mark in terms of resilience and efficiency, while helping us to optimise the fairly confined space in our data centre.” Ultra-efficient cooling is provided by a combination of external chillers and condensers located on the roof of the building, in addition to the InRow DX systems deployed

within the Pod. The facility is also managed using Schneider Electric’s StruxureWare for Data Centers DCIM (Data Centre Infrastructure Management) software, part of the EcoStruxure for Data Centers Solution. “As a team we knew we could run an on-premise data centre cost-effectively,” Wren continues, “but the Schneider Electric infrastructure components we have selected have been key to that process, ensuring the capacity is utilised in the most optimum way.” As well as provisioning and installing much of the infrastructure equipment in the data centre, Comtec Power continue to provide monitoring and maintenance support in collaboration with Angel Trains’ IT staff. Ongoing support provided by Comtec includes taking responsibility for rapid response to any faults in the infrastructure equipment, such as failures in air-conditioning units, including fans, and UPS battery malfunctions. “Our team can handle some of the smaller tasks internally,” continues Wren, “but under our maintenance service agreement, Comtec can proactively monitor and react to any faults within our core data infrastructure.” As part of a recent upgrade to the standard maintenance agreement, Angel Trains has recently connected the data centre infrastructure compnents to Schneider Electric’s EcoStruxure IT monitoring solution, which was previously known as StruxureOn. This delivers detailed 24/7 monitoring and critical insights straight to the users’ mobile phone as well as Comtec’s engineering team. “Through the remote diagnostics, Comtec can engage quickly to begin fixing issues while proactively avoiding any serious situations or downtime from developing,” Wren concludes. “We chose Comtec because they have the most experience. They built the system and we are comfortable operating in partnership with them as our trusted advisors.” l August 2018 MCP

INDUSTRIAL INTERNET OF THINGS

available much earlier when designing facilities, for testing and trials. The panel explained that these micro data centres are being designed for harsh industrial environments, while UPS and backup generators are also being tailored to the manufacturing process.

How is the data centre responding to IIoT demands? The rapidly developing edge data centre industry has an important role to play in the manufacturing sector, as production lines become increasingly reliant on IT. Louise Frampton reports on a recent DCD Debate in which leading experts highlighted the challenges

esilience is becoming increasingly important for the manufacturing sector, as digitisation has become a business priority for many organisations. “Manufacturing has changed dramatically. Today, production lines cannot run without IT; and, without electricity, the IT will not run. We have to keep things working,” commented Mark Howell, global IT facilities planning and engineering, at Ford Motor Company. “Older factories need to adapt – infrastructure varies from being very advanced to being fairly old and this includes the electrical distribution and IT infrastructure. Making this resilient and keeping the production lines running is something that many companies are grappling with.” Arup UK advanced manufacturing leader Mark Bartlett pointed out that a huge amount of IT is being introduced to support advanced MCP August 2018

manufacturing models, not only in the UK but also overseas. “There is a trend for Industry 4.0 and Industry 5.0 in new facilities and we are seeing this being linked with IoT,” he said. Bucking the Cloud trend While there is widespread migration to the Cloud in general, the manufacturing sector is bucking this trend, according to the panel. “Outsourcing applications to the Cloud isn’t really conducive to the manufacturing process,” explained Victor Avelar, director and senior research analyst, Data Centre Science Centre, Schneider Electric. “We are seeing more on prem IT kit. In particular, servers are being placed in micro data centres – small racks filled with specific IT solutions that control a particular portion of the manufacturing process…” Avelar added that there is an even greater need for cooling and powering of this equipment.

“The more it becomes tied into the operations technology – the robots, the machine, the conveyor belt technology – the more critical the IT becomes,” he said. “Every machine is computer controlled… and these machines are connected to other machines. They need to be sequenced and running exactly at the right time. All that computing cannot be up in the cloud. We are having to put computing onsite and make it more robust.” To make the best use of the data and analytics to increase productivity and manage costs, many applications need to be run at the edge, very close to the load and this is being driven by bandwidth and latency considerations. Not only is manufacturing becoming ‘just in time’, but IT infrastructure is also being constructed in a similar manner. Bartlett commented that Arup is witnessing increasing use of small, modular data centres and they need to be made

Critical thinking “There is a huge criticality in this space,” continued Avelar. “As IT is becoming critical to the manufacturing process, you need to take steps to ensure reliability is maintained. Where medium voltage machines are being used, active filters may be required; you will need to ensure there are no voltage spikes. You also need the IT equipment to be backed up – we are seeing the increasing use of lithium-ion batteries and singlephase UPS. Compared to valveregulated lead-acid batteries, they have a much longer lifetime and are better suited to harsh environments. In terms of technology, liquid cooled servers are also starting to make a resurgence. This could be a potential option in the future, in these harsh environments,” he commented. Bartlett added that Arup is also seeing more energy monitoring at a machine level but facilities also want to have the ability to look at the quality of the power supply and investigate interruptions and spikes. “Manufacturers want a lot more data and analytics,” he observed. Manufacturers such as Schneider Electric are driving down losses in terms of their UPSs but with efficiency already at 97%, and 99% in conversion mode, there is a limit to how much further this can go, according to Avelar. “In the future, we are going to see more interaction with the grid. It will be interesting to see where manufacturing heads with this; there is potential to use energy storage – manufacturers could participate in peak shaving and be financially rewarded by the utility," he concluded. l

To view the webinar, visit: https://tinyurl.com/ya58g5v8 missioncriticalpower.uk

INDUSTRIAL INTERNET OF THINGS

With the increasing adoption of edgebased systems and Internet of Things technologies, there is a growing need for simplified infrastructure that can be remotely managed. Louise Frampton reports

or organisations with critical business applications, each minute of unplanned downtime can have severe repercussions on the company – from lost revenue, to not meeting service level agreements (SLAs) and brand reputation damage. This exposure to risk is a continual concern to CIOs and COOs, and is a key reason why more IT decision makers are looking to implement fault-tolerant or high availability solutions. In fact, 38% of respondents to a survey undertaken by Enterprise Strategy Group (ESG) on behalf of Stratus Technologies said they plan to increase the number of production platforms protected by always-on infrastructure in the immediate future. “Unplanned downtime continues to be a huge vulnerability in today’s IT systems and, alarmingly, the vast majority of companies are not even tracking downtime with a quantified measure of cost,” says Jason Andersen, vice-president of business line management at Stratus Technologies. “We believe that with the increasing adoption of edgebased systems, including Industrial Internet of Things [IIoT] technologies, the costs and risks associated with downtime will continue to trend upward.” This is going to become a significant issue in the near future. Research from Gartner predicts that, by 2022, more than 50% of IIoT analytics MCP August 2018

Reducing unplanned downtime at the edge will be performed at the edge, compared with less than 10% today. The promise of edge computing in industrial environments means getting the right device data in near real time to drive better decisions, and even control industrial processes. However, for this to work, the edge device, its embedded software, edge servers, the gateways and Cloud infrastructure need to be operating correctly at all times. In today’s industrial environments, companies are faced with mounting pressure to be more efficient with fewer resources. At the same time, industrial technologies are becoming more powerful and more connected, yet are being deployed where there are little or no IT resources. The influx of data from these technologies and devices is driving the need for – and growth of – new edge

computing infrastructure. According to research from IDC, revenue of edge computing infrastructure will reach $3.4bn (£2.58bn) in 2021, growing at a CAGR of 22%, with IoT and IT/OT convergence driving most of this opportunity. With this growth comes the need for easier ways to ensure the availability of applications that run critical operations at the edge and collect and process this data. In fact, in a recent market report by ARC Advisory Group, more than 90% of those surveyed indicated that, as edge computing grows, organisations will need a simplified edge infrastructure that can be remotely managed. Virtualised computing at the industrial edge Stratus Technologies, which specialises in continuous

availability solutions for mission-critical applications, has developed ztC Edge to address this growing need. A zero-touch, fully virtualised and self-protecting computing platform, the technology is specifically designed for industrial edge environments. ZtC Edge is designed to reduce the IT burden for virtualised computing at the edge and can be installed by an end user, with no prior knowledge of the system, in as little as 30 minutes. Greg Hookings, head of business development – IIoT, Stratus Technologies, comments: “Meshing information technology and operational technology is a challenge for industry. It is not simply a case of ‘sandwiching’ the Cloud onto your operational infrastructure. At the industrial edge, we are taking information technology facets and missioncriticalpower.uk

Unplanned downtime continues to be a huge vulnerability in today’s IT systems and, alarmingly, the vast majority of companies are not even tracking downtime with a quantified measure of cost

industrial or IIoT applications. Stratus ztC Edge is comprised of redundant nodes that act as a single system. With live migration of virtual machines, data replication, and redundant networking, it instantly provides protection for applications and data. The system continually monitors itself, evaluating its health and performance. If ztC Edge detects a potential central processing unit (CPU) or memory failure in one node, it proactively moves virtual machines from one node to the other, ensuring application continuity. Its selfprotecting and self-monitoring features help reduce unplanned downtime, making it suitable for unmanned stations and remote locations. Doosan: fuel cell technology These features make ztC Edge an ideal solution for Doosan, a provider of sustainable energy solutions. The company is using the technology to ensure uptime for remote power solutions around the world. Offering 98% availability, Doosan’s hydrogen fuel cells

operational technology facets and creating a new technology layer. We need to get away from the complexity of IT.” Key features include: • Built-in virtualisation for up to three virtual machines, each running different industrial control or IoT applications • Automated application and data recovery • Automated site recovery •Rugged, hot-swappable nodes that can be wall- or DIN railmounted • Cloud-based health monitoring and built-in remote management services Through pre-installed virtualisation software, and user-friendly configuration tools, ztC Edge simplifies the process, and shortens the time it takes to get critical applications up and running. It can manage up to three virtual machines, each running different missioncriticalpower.uk

Stratus ztC Edge

Doosan hydrogen fuel cell

deliver more than 440kW of electricity and 450kW of heat, have a low carbon footprint and near zero, ultralow emissions. The fuel cells operate without combustion, so are virtually pollution free. Since the fuel is converted directly to electricity and heat, a fuel cell’s total system efficiency is far higher than conventional power generation equipment. The fuel cells provide environmentallyfriendly backup power and heating for a wide variety of applications – including data centres, hospitals and utilities – and are ideal for applications where traditional renewable technology is unfeasible. Uptime for the Doosan PureCell units is ensured through the application of FacilityConneX software, designed to deliver energy savings and operational efficiency through smart asset monitoring, diagnostics and advanced predictive analytics. The software is claimed to reduce costs by up to 35%. FacilityConneX CEO Tom Schiller explains: “Stratus ztC Edge gives us a reliable and robust server to connect to the FacilityConneX Cloud platform, where we are monitoring Doosan’s fuel cells around the world. It provides us with a versatile platform to run our GE Digital iFIX and Historian software, acting as HMI/Scada as well as data collection and IIoT server.” Hookings adds: “The hydrogen cells need to be maintained and replaced at intervals and the analytics allow personnel to be deployed when the requirement arises; ztC Edge is all about providing server redundancy in a remote industrial environment. “In the past, an application such as this would require PLCs and some kind of IT device running the Scada software. We have been able to reduce a lot of the accessories, previously required, and replace this with an IT solution. This reduces the complexity of the application and significantly reduces cost.” l August 2018 MCP

DEMAND-SIDE RESPONSE

Sky’s the limit for DSR revenues? Is demand-side response an attractive proposition and are mission critical sites now coming on board?

emand-side response is becoming more critical as a method to balance the grid with increasing amounts of intermittent generation and a reluctance to have fossil fuel spinning reserve in place. Sites with back-up generation are ideally suited to this and therefore critical facilities, such as hospitals, data centres, utilities. airports and the like, have the opportunity to make money and help to balance the electricity system. While mission critical sites are increasingly coming on

board, there is huge untapped potential across a variety of sectors. It is clear some are more open to adopting DSR than others. Data centres, in particular, have required some convincing. Equinix, a colocation data centre provider believes that DSR is an exciting area but the conditions will need to be right for the company to consider making the next step. Michael Winterson, managing director of Equinix Services, comments: “Governments want digitisation as it is the only way to sustain

a growing and more urban population. If we can digitise what were once physical services, the planet will benefit but we need governments to help us. One of the most interesting concepts we have been discussing is: ‘What if the data centre augmented its battery facilities, so that it could enter into contracts with the government and allow the government’s grid to start micromanaging us?’ We could come off grid or extract extra energy but this requires joined up thinking. It requires long-

term planning and we need to engage national government departments. “Firstly, participation has to make sense for national shareholders, so there must be an economic benefit for investing in the technology; secondly, participation will make sense if customers demand it; and thirdly, it will make sense if the government wants to make it happen and treats it as being for the national good, forcing the economy to deal with it. “At the moment, it doesn’t make economic sense for us to do this. The offer on the table compared to the capital that we would have to spend, means that it wouldn’t pay back in any decent response time.” So could government regulation be the answer to increasing uptake? Winterson believes this isn’t the answer. “If government regulated what we did it, they would impose a burden and a cost on our business that would mean a worse return for shareholders, so our stock price would go down and we would deliver less of a service to the economy, or we would have to pass on the cost to our clients. “This would mean that businesses would keep their ‘dirty’ servers in their own ‘dirty data centres’ in the basement, as they couldn’t afford to migrate to a large, green data centre,

DSR Event: panel experts to discuss opportunities Visitors to The Energyst’s DSR Event can gain key insights into the challenges, solutions and requirements for particiption in grid balancing schemes. Discussion topics at the DSR Event will include: • DSR market overview: What’s changed and where next? National Grid details changes to its balancing services and what that means for DSR providers • Generation versus load: Impact of recent policy and regulatory changes. What does the future hold for generation forms of DSR in light of incoming rule changes, and what opportunities does this open up for ‘load’ providers? • Balancing risk and reward: Are businesses comfortable sharing more risk to unlock higher revenue? What challenges do end user organisations face in unlocking flexibility, and is there appetite for increased risk to make business case stack-up? • Follow the money: Where’s it going, how can you access best value? How businesses are mitigating revenue cannibalisation in contracted markets with more agile strategies • Hybrid DSR and storage: Less risk, more reward? How businesses can reduce risk and maximise benefits by pairing batteries with DSR assets Last year’s event was oversubscribed. Register in advance for the 2018 DSR Event to avoid disappointment. The DSR Event 2018 will take place on Thursday 13 September at the prestigious Banking Hall, central London. For further details and to register, visit: https://dsrevent.uk

MCP August 2018

missioncriticalpower.uk

Sponsored column

like Equinix. People have to understand the negativity of regulation. We need to think big about this; about the future of a digital economy,” Winterson argues. Other data centres are already pushing forward into grid balancing schemes and report that it makes sense for their individual business model. Some data centre operators go as far as saying that the resilience of their data centre is improved by taking part in DSR schemes. Ari Kurvi, data centre manager for Yandex Oy’s facility, based in Mäntsälä, Finland, comments: “The only way you can ensure your system works is to test it end-to-end. We have seen no increase in risk by participating in DSR – instead, we have reduced our risk by testing our systems more frequently and more thoroughly. We are ahead of any disaster that may happen on the grid and can react much quicker to

revenue and build operational resilience from its participation in both schemes, while actively contributing to EirGrid’s system stability as they integrate more renewables on the grid. EirGrid’s DS3 scheme is a new ancillary services programme that will provide incentive payments to large energy users that can enact fast-acting distributed energy assets when system frequency drops to a certain rate. Prior to the May 2018 launch of the DS3 programme, EnerNOC was already a provider of fast responding aggregated grid services in Ireland as part of an EirGrid pilot project. Ian Clarke, energy manager, Dublin Airport, comments: “Participating in EirGrid’s latest grid balancing schemes aligns perfectly with our asset management and energy objectives. We are delighted to become part of the EnerNOC network of businesses that

Participation has to make sense for national shareholders, so there must be an economic benefit for investing in the technology any frequency drop compared to other data centres that are not participating in DSR schemes.” Dublin Airport Authority (DAA) is also among the increasing number of critical sites starting to come on board with grid balancing schemes. DAA is partnering with EnerNOC to help deliver more than 11MW of flexible capacity to both EirGrid’s recently launched ancillary services programme and the future capacity market due to launch later this year. DAA is already providing more than 3MW of very fast responding flexible capacity within the DS3 (Tertiary Operating Reserve 1, or TOR1) scheme, and will also enrol more than 8MW of flexible capacity into the new Integrated Single Electricity Market (I-SEM) for Ireland and Northern Ireland, due to launch 1 October. DAA will earn streams of missioncriticalpower.uk

are enabling Ireland’s grid resilience while furthering our own operational resilience. Our enrolment, testing, and early participation has been simple and seamless.” EnerNOC’s Sam Scuilli, will be joining the panel discussion on ‘Capacity Market Changes and Opportunities for Businesses’ at the DSR Event. The event will provide valuable insights from experts, aggregators and energy users. National Grid needs more businesses to reduce power use, or switch to on-site generation instead of asking power stations to ramp up at times of system stress. Ultimately, there are significant opportunities for businesses to avoid costs and generate revenue. The Demand-Side Response Event, hosted by Mission Critical Power’s sister title, The Energyst, will outline these opportunities and how to take part. l

Avoid paying the price for poor power quality ABB’s power quality manager, Steve Joyce, explains how many businesses in the UK have experienced a sudden jump in energy bills following the introduction of the DCP161 legislation by Ofgem – and what they can do to work around the legislation. New legislation introduced in April by energy regulator Ofgem has been affecting businesses. Since DCP161 was introduced, businesses can be charged a penalty tariff when they exceed their maximum half-hourly (HH) metered electricity consumption. Previously, any excess energy was billed at the standard rate but utilities can now charge up to three times the standard rate depending on the location and the voltage level. The new rule was introduced after network operators reported the need to make major investments in upgrading grid infrastructure to meet growing peak-time demand. Ofgem ruled that these costs should be met by the customers that draw more than the maximum agreed in their contracts. As a result, some businesses have seen a step change in their bills. However, there are options available for businesses that want to take control and avoid the penalties by preventing their consumption going over their maximum HH value. One option that has a lot of potential is to use energy more efficiently by improving Power Factor (PF). PF is the ratio of useful energy to total energy consumed and is therefore a measure of how efficiently a site uses electricity. Low PF affects industrial and

business customers that operate a lot of AC motors, arc welders, furnaces and fluorescent lighting. These store energy briefly, causing electricity to be consumed as ‘active power’ that does the work and ‘reactive power’ that goes to waste. In layman’s terms, active and reactive power can be compared with a frothy latte. Active power is like the main body of the coffee, whereas reactive power is like the froth, filling space in the glass. A little froth is good but too much is a waste. Ideally, PF should be between 0.95 and 1. Below 0.95, you should take action by installing Power Factor Correction (PFC) equipment. This ensures that the site uses only active power, eliminating reactive power and reducing overall energy consumption. As a result, installing PFC equipment can help a business reduce overall energy consumption and stay within its HH consumption target, avoiding penalty fees under DCP161. Payback for PFC equipment is typically less than two years. Many businesses can simply check PF as it is listed on some energy bills. However, for those who don’t, we have a team of surveyors who can carry out a free, no-obligation survey for customers throughout the UK. If your PF has dropped below 0.95, we can then work with you to design and install a PFC solution. This will deliver real savings on electricity bills from day one and help to avoid excess charges under the DCP161 scheme. Learn more and request a free on-site power factor survey at new.abb.com/uk/power-factorcorrection

DEMAND-SIDE RESPONSE

s the heatwave spreads across the country, and we are all reaching for glasses of cool water, hoping for water suppliers to meet the increasing demand and provide an unlimited stream straight to our taps, it is a good opportunity to notice how much the safety and comfort of our lifestyles rely on the resilience of mission critical businesses in the water sector. Regardless of external circumstances, whether it is a long spell of hot weather or an electricity failure, mission critical sectors need to be resilient and capable of delivering services. “The priority for mission critical businesses, whether they are hospitals, data centres, banks, utilities or transport networks, is to secure the undisturbed continuity of their operation. Understandably, their primary focus is to predict and minimise any risks. That is why it is vital for the businesses that rely on critical power to take more responsibility for their energy management by gaining a better understanding of dynamic changes that occur in the energy market, its challenges and opportunities,” says Michael Phelan, chief executive at GridBeyond, an energy technology and DSR provider. Mission critical businesses are gaining a deeper understanding of demandside response programmes, embedded generation and onsite energy storage. As such, many are starting to realise that participation in DSR not only enables them to generate savings and revenue while simultaneously increasing green credentials but also strengthen a site’s resilience and help manage risk. “At GridBeyond, all of our clients’ sites and assets participate in DSR via the technologically advanced platform, powered by AI and machine learning algorithms. The platform, connected to highly sensitive and accurate sensors, enables precise monitoring of energy MCP August 2018

Opening the water industry flood gates There is increasing uptake of demand-side response in the mission critical water industry. Irish Water and Northern Ireland Water are paving the way by venturing into an area many have been too nervous to approach

performance within each asset and helps to identify any inaccuracies in their operations before a minor issue becomes a significant problem that might endanger the business operations,” Phelan explains further. As the UK and Ireland rapidly decarbonise to meet CO2 reduction targets required by international law, mission critical I&C businesses help to integrate more renewable energy sources onto the electricity networks, promote sustainable manufacturing and improve their environmental

credentials by participating in DSR. Ireland aims to reach its energy target of 40% renewables contribution to gross electricity consumption by 2020. However, as the renewable energy production tends to be volatile, EirGrid needs support from large energy users such as Irish Water or Northern Ireland Water to maintain grid frequency. “For our Irish clients in the critical water sector, environmental factors were the main reasons to partner with us in delivering DSR services to

EirGrid. Since last year, we have been supporting Irish Water in meeting both its environmental and business objectives, and now we are delighted to announce our cooperation with Northern Ireland Water. This makes GridBeyond the only energy technology and DSR provider to work with all the main water services providers across the full island of Ireland,” says Phelan. Irish Water The opportunities to benefit from the smart grid were identified by Irish Water as part of its energy management missioncriticalpower.uk

43 Denis McGuire, process optimisation manager at Irish Water, says: “Through our involvement in EirGrid’s programmes we are increasing our business efficiency, contributing to decarbonisation of electricity production and helping the country to meet its environmental targets.” Northern Ireland Water Northern Ireland Water provides water and wastewater services for almost 1.8 million people in Northern Ireland – producing 570 million litres of clean water per day and safely recycling 340 million litres of wastewater back into the environment. Uniquely placed NI Water, a state-owned company that is publicly funded, has recently celebrated a decade of delivery with substantial customer service and efficiency improvements. NI Water operates £3bn worth of assets. The energy required to deliver its services makes NI Water the single largest electricity consumer in Northern Ireland. Jane Mellor, NI Water’s

tendering process, GridBeyond has been awarded a five-year contract to utilise NI Water’s assets to provide Capacity and DS3 system services. “NI Water’s participation in the Capacity and DS3 System Services Market is tangible evidence of our commitment to meeting environmental targets, by facilitating efficiencies and increased renewable generation across the electricity grid to both reduce costs and CO2 emissions,” says Mellor. “By working with GridBeyond, an experienced energy technology company, NI Water is demonstrating a continuing commitment to delivering high quality services, while simultaneously enhancing natural and social capital.” Paving the way In summary, Irish Water and Northern Ireland Water are paving the way for mission critical sites. By venturing into an area many have been nervous to approach, they are uncovering numerous ways to boost their resilience.

Through our involvement in EirGrid’s programmes we are increasing our business efficiency, contributing to decarbonisation of electricity production and helping the country to meet its environmental targets initiatives to improve energy performance by 33% by 2021 and to achieve energy cost savings across the business. Paul Byrne, electricity management analyst at Irish Water, explains: “We recognise that a smart electricity grid is the future for electricity generation, transmission, and consumption in Ireland. As a company, we are committed to supporting environmental sustainability and helping in incorporating more renewable energy onto the grid by providing balancing services to match electricity demand missioncriticalpower.uk

to the supply on the whole island.” Last year, in conjunction with 10 local authorities from across the country, and with the support from GridBeyond, Irish Water commenced the first phase of participation across its largest energy consuming sites. Currently, 23 of Irish Water’s sites are connected to GridBeyond’s platform and participate in demand-side response programmes, assisting the balancing of the grid by turning down, and to the lesser extent turning up, the power consumption.

head of operational procurement, says: “We consider sustainability and climate change mitigation as priorities that inform our decisions on the future direction of the business. NI Water is committed to using innovative approaches to energy management and new technologies to deliver water and wastewater services for the lowest financial and environmental cost, while simultaneously maximising consumer and community benefit.” During the recent competitive

First, by assisting the grid at times of peak demand, thereby improving overall grid resilience. Second, by monitoring and controlling their energy demand, the machine-learning technology recognises patterns in consumption both onsite, and by using industrial benchmarks for performance. When an anomaly occurs an alert is raised, providing the information needed for predictive maintenance. Through such measures, a fault can be detected well in advance of failure. l August 2018 MCP

DATA CENTRE OPTIMISATION

ABB digs deep to power Lefdal ahead There was a time when the mineral olivine used to be excavated for industrial purposes, such as making steel. Now, that steel is coming back underground to house one of Europe’s largest, and greenest, data centres

he Lefdal Mine Data Centre is built 150 metres into a mountain in what was formerly an underground mine for excavating olivine – also known as the gemstone peridot – a green, high densitymineral used in steel production. Located on Norway’s west coast, between Måløy and Nordfjordeid, the six-storey mountain hall facility is powered exclusively by renewable energy produced locally, while being cooled by water from a nearby fjord. ABB has supplied the critical power infrastructure, which provides clean energy generated by four glacial hydropower stations and

two windmill farms with a combined capacity in excess of 300MW. Data centres are among the biggest consumers of energy, yet Lefdal Mine is remarkably energy efficient because it uses cold water from the 565m-deep fjord as a coolant. The data centre is located below sea level, eliminating the need for expensive highcapacity pumps to lift the fjord’s water to the cooling system’s heat exchangers. The result is that the cooling solution will have power usage effectiveness (PUE) of between 1.08 and 1.15 for 5kW rack, making it among the greenest data centres in the world

with 30-40% energy savings compared with traditional facilities. Lefdal Mine Data Centre marketing director Mats Andersson says: “Cooling is crucial, because these servers generate huge amounts of heat. Because water cooling is so efficient, these server containers can run up to 50kW of power, where you would normally expect just 7-8 kW with traditional air cooling.” Maintaining secure operations 24 hours a day is crucial, with redundant systems in place to ensure the data centre is always operational. To meet the powering

Cool savings from Russia, with IXcelleration Cool technology is helping Moscow One data centre to halve energy use following its latest expansion MCP August 2018

ussian data centre provider IXcellerate has launched the third expansion of its Moscow One campus, creating one of the largest data halls in Russia. Vertiv has been a technology partner since 2013, providing

thermal management products, including refrigerant gas-based CRAC units, as well as free cooling chillers. For its latest phase of expansion, IXcellerate selected Liebert AFC adiabatic free cooling chillers. It also

deployed the Liebert PDX with Vertiv EconoPhase, an aircooled direct expansion system with a refrigerant pumping module to allow a shift from compressor to economiser mode when conditions suit. Cooling typically accounts missioncriticalpower.uk

ABB has supplied the critical power infrastructure, which provides clean energy generated by four glacial hydropower stations and two windmill farms with a combined capacity in excess of 300MW

Energy savings being delivered, compared with traditional data centres

everything starts with power. You need transformers, you need generators. So, based on the good relations we have, we started to discuss how to get ABB on board.” Lefdal Mine Data Centre is coming up at the right time – when Norway is aiming to become a superpower for green data centres. The nation has a surplus of renewable energy – 97% of electricity generated in the country comes from renewable sources, mainly hydropower, according to state-owned Innovation Norway. The nation’s solar sector – though still less than 1% of power generation – is growing fast, with panel installations

growing by more than 300% in 2016. The Norwegian government is looking to encourage more digital innovation to create new industries that create jobs and boost economic growth. In February 2018, the government released its data centre strategy, Powered by Nature, which stressed that attracting data centres and international investments is an important part of their industrial policy. With such incentives, and a fast-growing need for more data centres powered by renewable energy, Lefdal Mine will have an edge. As more of the world becomes digital, ABB will be powering Lefdal and Norway ahead. l

management to the provision of Liebert EXL S1 uninterruptible power supplies (UPS), which offer double conversion efficiency of up to 97% and up to 99% in eco mode, as well as optimised efficiency at partial load. The Liebert EXL S1 is also compatible with modern lithium-ion batteries. “Lithium-ion batteries used in UPS promise benefits

including significantly smaller and lighter footprints, faster charge/discharge rates, and higher energy density compared to traditional valve-regulated lead-acid (VRLA) batteries,” says Jeffrey Fidacaro, senior analyst at IT analyst 451 Research. “Deploying lithium-ion should allow IXcellerate to maximise its available white

space for customer equipment. IXcellerate will also benefit from lithium-ion’s longer battery life, ability to operate in a wider temperature range and lower maintenance relative to lead-acid. “451 Research expects these features could provide significant savings on cooling and an overall improved total cost of ownership.” l

challenges due to the physical size of the facility, ABB has built a medium-voltage backbone for the entire facility. To meet any emergency situation, ABB also provides a decentralised UPS system, which means that each section inside the data centre has its own UPS installation. If there is a problem with the grid, the UPS kicks in within a couple of milliseconds and ensures reliable power supply until the backup generators come online. Providing a powering system that can remain reliable as the centre grows (from the current 10MW to 200MW in the next three years) is of particular importance. When its growth is complete, Lefdal will be among

Europe’s largest data centres. About 120,000m2 (1.3 million sq ft) of white space is currently available in the data centre, much of it in containers shipped by Rittal and parked in the former underground mine passages. Rittal executive vicepresident Andreas Keiger says: “ABB was one of the first to be involved in the project, because

for 40% of energy consumption within data centres but this combination of technology allows up to a 50% reduction in energy consumption compared with a conventional air-cooled unit, and partial power usage effectiveness (pPUE) levels as low as 1.05. The equipment deployed for IXcellerate also extends beyond critical thermal missioncriticalpower.uk

40%

August 2018 MCP

CABLES & CONNECTIONS

it is vital to ensure that the chosen cables comply with CPR regulations to reduce the damage and prevent the spread if a fire ever did occur in a data centre.

What lies beneath the sheath? Julie Banks, UK sales manager at Nexans, discusses how it is everyone’s responsibility to establish if the cables they are using really are fit for purpose

ata centre managers face a number of challenges: to keep a reliable flow of information, to manage the consumption of energy; to plan future migration, to keep up with exponential data growth; and reduce running costs. Therefore, when choosing the right cables for data centres, they must understand what lays beneath the sheath. About 60% of cables used in the UK are currently imported and sold through distribution channels. This means that identifying the brand of a cable can be difficult as it is not always completely clear where a cable has come from. In addition to this, the increasing price pressure (partly due to the increase in the price of raw material, such as oil, ethylene, PVC and the volatile price of LME) and acceleration of commoditisation can cause buyers to seek better prices, which in turn can result in quality being put at risk. Safety starts with you Not all cables are manufactured to meet the required approvals for data centres. It is possible that some cables are designed using inferior methods and materials that will not meet

MCP August 2018

local requirements. If there is any doubt that a cable is not fit for purpose, the specification of the product should be checked and questioned. It is up to the customer to ensure that the cable they purchase meets the required safety standards for its application. To make an informed decision on the safety of a cable, it is important to ask yourself: “Do I really know what’s beneath the sheath?” New CPR regulations One way to prevent the use of sub-standard products in data centres is to only purchase those that comply with the Construction Products Regulations (CPR). Since 1 July 2017, every product must be CE marked (which means a product complies with all European directives and regulations) and needs to have a Declaration of Performance (DoP). CPR has been introduced by the European Commission to enforce that construction works are designed and executed so as not to endanger the safety of persons, domestic animals or property nor damage the environment. The aim of CPR is to create a harmonised set of rules for the marking of construction

The importance of CPD An excellent way to improve knowledge in the safety of cables is through experience, learning and application in the workplace. Many manufacturers in the industry offer various CPD modules that can help to bring a proactive approach to learning. The modules combine all sorts of training from workshops, e-learning, best practices and classroom education. Consider asking the manufacturer for more information on the cable, even if the end user does not have a direct relationship with the manufacturer. The manufacturer has the technical products within the EU. This expertise, knowledge and will provide a common technical understanding to support the language in order to assess the entire channel. They have performance of construction the responsibility to provide products and ensure the accurate and clear technical availability of information, so information to help customers comparisons can be made. make more informed decisions. Although it is the primary Other bodies can also support responsibility of the in making informed decisions manufacturer or importer to such as the Approved Cables ensure that products placed on Initiative (ACI) and British the EU market comply with Cables Association (BCA). CPR obligations, the The BCA is the UK installer should also trade association for check that the manufacturers of cables comply insulated metallic by requesting a and fibre optic DoP. cables, wires and By doing this their accessories. It and becoming provides the latest educated in this information about the area, the installer industry and contributes can be reassured that to the development of the product meets British, European and CPR, as all drums and Julie Banks International Standards. labelling should have It also helps publish codes of the same reference number as practice and guides to safe the DoP and the same CE mark manufacturing usage. and the year in which the CE marking was first affixed. The responsibility is yours Fire safety is an issue across Ultimately, it is everyone’s multiple sectors, including responsibility to ensure that IT. In fact, according to a cables used on any project really Capitoline survey, 21% of all are fit for purpose – from the non-IT catastrophic failures of manufacturer, to the distributor, data centres are due to fire and to the installer and finally the 5% are due to a malfunction end user. l of the fire system. This is why missioncriticalpower.uk

PRODUCTS

Self-addressing energy transducer range Carlo Gavazzi has launched its latest selfaddressing DIN rail mounting energy transducer, ET272, for the control of electrical parameters in critical load applications such as datacentres or applications requiring detailed monitoring that can reach every single branch of the electrical distribution system. The EM272 measures up to two threephase loads or up to six single-phase loads simultaneously. By simply plugging in the connector, auto connection of the primary current via the external miniature split core current sensors provides zero wiring errors. The meter is automatically programmed for CT ratio accuracy and calibration. A selection of single-phase variables such as V kW, kVAR, PF and total energy measurement kWh and kvarh (imported/

Robust and compact UPS takes it nice and Easy Schneider Electric has announced the launch of Easy UPS 3S in Europe, Middle East, Africa and Latin America. Easy UPS 3S is a 10-to-40kVA uninterruptible power supply (UPS) that offers robust electrical specifications and compact design, and is easy to install, easy to use and easy to service – making business continuity easy for small and medium businesses. The Easy UPS 3S is a combination of an optimised footprint design and advanced product features, protecting critical equipment in many environments from damage due to power outages, surges and spikes. This unit delivers up to 96% efficiency in double conversion mode and up to 99% efficiency in energy-saving ECO Mode, taking the industry standards for the 10-to-40kVA UPS to a new level. “Easy UPS 3S addresses a market need we see for a solution missioncriticalpower.uk

that offers best-inclass power availability, reliability, manageability, quality, and convenience for small and medium businesses, data centres, and manufacturing facilities,” said Pedro Robredo, vice-president, Secure Power, Schneider Electric. “We look forward to bringing this new threephase 10-to-40kVA offer to EMEA European, Middle Eastern, African and Latin American countries. With its easiness, competitiveness, and versatility, Easy UPS 3S is ideally suited for use in a wide range of industries and environments.” This new UPS offer is rugged, with a wide input voltage window and strong overload protection all in a compact and lightweight footprint. It has been designed and tested following rigorous procedures – making it an ideal fit in both the data centre space and in light industrial environments.

exported) can be taken and offers instantaneous variables readout of three digits or energy meter readout of 6+1 digits. The EM272 and WM50 family suit critical load applications or where controlling multiple loads is vital such as in data centres, hospitals and universities where high reliability and extreme measurement precision is required for monitoring of any single branch of the distribution system. Operating temperature range is from -25°C to +55°C and the front panel offers IP40 protection and meets the metrology standards Class 1 (kWh) of EN62053-21 and Class 2 (kvarh) of EN62053-23. As with other Carlo Gavazzi energy meters, power analysers and transducers, the ET272 carries CE marking and UL certifications.

Flexible cabinet design Panduit, a provider of leadingedge physical infrastructure and connectivity solutions, has announced the availability of Net-Verse Cabinets in Europe. Net-Verse Cabinets are a highly flexible cabinet design for applications that require scalability, security and cost efficiency in on premises and colocation/cloud data centre or enterprise deployments. Adaptive design features include fully adjustable equipment rails that cater to both server and switch applications. The cabinets have enhanced physical security features such as multi-point locking front and rear doors and locking side panels. For high density compute environments, its versatility is further supported by two full-length combination PDU/ lashing brackets that can accommodate up to two PDUs, including Panduit’s recently launched Smartzone G5 Intelligent PDU, on each side of the cabinet. Sander Kaempfer, business development manager, data centre, Panduit EMEA, said: “Panduit’s Net-Verse Cabinets offer customers a flexible and cost-efficient cabinet solution for enhanced data centre agility and scalability, especially

in the colocation and cloud environments.” The new Net-Verse Cabinets are designed for applications that require optimal thermal management within the latest containment environments. Net-Verse Cabinet front and rear doors provide a generous 69% opening for optimised airflow.

August 2018 MCP

PRODUCTS

High-capacity power testing solution Global loadbank manufacturer Crestchic has expanded its portfolio of containerised loadbanks with the introduction of a larger 30ft unit – the largest container solution to be offered by the company to date. Ideal for facilitating high capacity power testing and supply from one container, the larger units can house loadbanks at 380V-690V up to 8000kVA resistive-reactive (0.8pf ) and 8000kW resistive-only (1.0pf ). Also available are medium voltage units of 4000kVA at voltages up to 36kV.

Delivering a lower carbon footprint for expansive or high-demand environments, Crestchic’s new 30ft containerised loadbank offering has already proven to be the right solution for a number of high-profile customers. Most recently CSBS Corporation, Taiwan’s largest ship builder, purchased the larger sized containerised loadbank for use during ship building and repair, to allow for testing of the onboard generator sets, and to enable multiple generators to be paralleled on a common busbar. Paul Brickman, sales and marketing director at Crestchic, added: “Our new larger 30ft containerised loadbanks represent value for customers in need of substantial powertesting capacity and a multivoltage, multi-frequency supply from one unit. These units are easy to transport and deliver tangible energy savings and economies of scale. “The new sized units are already proving a popular choice for the ship building industry as well as the data centre and generator manufacturing markets.”

Cable organiser saves time and space Modern network infrastructures are reliant on optimum performance and maximum flexibility with their cabling. The Network Cable Organiser (NCO) from Rittal is ideal for ensuring this. It optimises patching work on the enclosure, with eight times faster assembly. This not only nets time savings but also creates space and order in the rack. The NCO will mean simple, fast and secure patching while surplus cable lengths, whether CAT 6 or fibre-optic, are simply withdrawn with a spring balancer. The NCO is a 482.6 mm (19”) cable storage system. It takes up 1 U in the network enclosure and contains 24 tested CAT 6, Class E patch cables or fibre-optic cables, each with a length of 1.6 m (sufficient for 23 U). The NCO has a modular structure and is made up of individual cassettes. The pulley system integrated into the cassettes allows surplus cable lengths to be drawn-in automatically. This ensures that every cable is available in the perfect length, eliminating the need to order and stock a variety of different cable lengths. The result is permanently well-organised cable management, providing a perfect overview of what is connected where. Another advantage offered by the NCO is its more efficient energy management. Avoiding airflow blockages that can arise as a result of surplus lengths of the individual cables inside the enclosure facilitates more efficient cooling. The NCO also saves on space because it only requires 1 U per switch. Cable management panels are not required, which in turn saves additional space. Added to which, the flat cables used in the NCO only use one-third of the space of a round cable. The flat cables still offer the same level of reliability, are 100% tested and the fibre-optic cables additionally have a measurement record on the cassette

MCP August 2018

Cable innovation Cable manufacturer Prysmian has announced two enhancements to its line of FlexRibbon products. This new line of products, which feature MassLink with FlexRibbon Technology, will support largescale data centres operated by major companies worldwide. Designed to maximise fibre density and duct space utilisation, Prysmian’s MassLinkwith FlexRibbon Technology products compact the maximum fibre count into the smallest cable possible by using extremely flexible fibre ribbons that can be rolled up for high packing densities or laid flat for ribbon splicing. The new 1728 and 3456 fibre count FlexRibbon products feature an ultra-compact outside plant cable design that contains bend insensitive fibres, small enough to fit into a 31.75mm and 38.1mm duct, respectively. The significantly smaller diameter and lighter weight cables are designed to offer optimal kink resistance and increased flexibility, which allows for easier installation and the use of smaller ducts. In the case of the 1728 fibre design, there is a 21% smaller diameter (38% volume reduction) compared with traditional flat ribbon designs. While the new FlexRibbon products provide high packing density, they also feature 200 and 250-micron fibre ribbons that still provide the advantages of mass fusion splicing. Philippe Vanhille, Prysmian Group senior vice-president, Telecom commented: “As the demand for more powerful, higher-density cables continues to increase, this latest offering will be a key asset for large companies to navigate underground duct networks.” Toni Bosch, Prysmian Group vice-president, Telecom Solutions added: “We began development of our FlexRibbon Technology product line to meet the needs of our main customers in North America requiring high fibre counts and fibre density, where ultra-compact design and efficient installation are necessities. “We have continued to heavily invest in the R&D of FlexRibbon Technology, employing the most advanced equipment, which has led these new FlexRibbon product offerings.” missioncriticalpower.uk

PRODUCT & SERVICES DIRECTORY

Contact sales@energystmedia.com BATTERY MANAGEMENT

FLOW METERING

GENSET CONTROLLERS

www.janitza.com

Clamp-on flow & heat/energy metering solution from Micronics.

DSE8610 MKII SHAPING THE FUTURE OF SYNCHRONISING.

www.micronicsflowmeters.com or call

POWERING MONITORING

+44 (0) 1628 810456

3 in1

Monitoring System

Reliable and efficient power supply

AVAILABLE

NOW



EnMS – Energy Management (ISO 50001)



PQ – Power Quality (EN 50160)



RCM – Residual Current Monitoring Video

REDUNDANT MSC

MADE IN BRITAIN

EXTENDED PLC ENHANCED FUNCTIONALITY COMMUNICATIONS

TO LEARN HOW DSE SYNCHRONISING SOLUTIONS WILL ENHANCE YOUR MULTI-SET APPLICATIONS VISIT WWW.DEEPSEAPLC.COM

T +44 (0) 1723 890099 E sales@deepseaplc.com

UPS 0999_Micronics U1000HM 44 x 110mm Ad05/10/2015 v4.indd 1 19:40

Premium Power Protection for Data Centres

Call: 0800 269 394

To feature your company’s products or services on this page contact

www.riello-ups.co.uk

sales@energystmedia.com

Reliable power for a sustainable world For further information, contact us on 0845 602 9471 or sales-ups.uk@ge.com

missioncriticalpower.uk

August 2018 MCP

Q&A

Andrew Stevens CNet’s CEO discusses being a ‘proper’ rock star, the lost city of Atlantis, and how energy consmers can earn money through simple energy savings, without having to pick-a-pocket or two... Who would you least like to share a lift with? Melania Trump… I wouldn’t be able to stop myself from saying: “Really, surely it can’t be the money?” You’re God for the day. What’s the first thing you do? I would rid the world of all types of addiction. This illness manifests itself in so many ways and causes so many unidentified issues. If you could travel back in time to a period in history, what would it be? The Romans. I would be fascinated to understand the drive and the engineering which led them to create so many amazing things.

fascinates me that so much evidence points to its actual existence. What would you take to a desert island? My dog, because he listens to me and seems to enjoy my endless ramblings, so he would be great company. What’s your favourite film? Oliver is by far my favourite. As a child it captured my imagination and I watch every year. My youngest son recently played Dodger in his school production. I went to see it four times.

Who are you enjoying listening to? I really enjoy most upbeat styles of music but in particular the Rolling Stones.

If you could perpetuate a myth about yourself, what would it be? It would have to be that Barcelona wanted to sign me as a 16-year-old footballer but I wanted to stay and play for my local Saturday team.

What unsolved mystery would you like the answers to? I would really like to know the truth about the Lost City of Atlantis. It

What would your super power be and why? I would want to be able swim and breath under water; it would be amazing to explore

I would have to be a rock star, and I mean a proper rock star… it would be great to travel the world doing something you love the oceans and explore places that no one has ever been. Imagine how good your holidays would be?

What irritates you the most in life? People who don’t make decisions and won’t take a risk.

What would you do with a million pounds? Probably the standard things but I would buy an old Triumph Stag in British Racing Green and a house in the New Forest as a retreat for all my family.

What should energy users be doing to help themselves in the current climate? Energy users need to understand what they are actually using and where the energy is being consumed and by who or what, then create a plan to manage it. Making simple savings first can then create the funds to deploy technology if required.

What’s your greatest extravagance? My watch. It was an impulse purchase. If you were blessed with any talent, what would your dream job be? I would have to be a rock star, and I mean a proper rock star… it would be great to travel the world doing something you love (sounds like my job actually).

One of the family: Oliver captured imagination as a child MCP August 2018

What is the best piece of advice ever been given? You are only as good as your last sale. It’s so true.

What’s the best thing – work wise – that you did recently? I took on an intern for the summer months. It has been great to see my team giving up their time to help him and to see how much he is learning. He is loving the opportunity and we may have just found a future team member. l missioncriticalpower.uk