Piecing together the R&D jigsaw  

THIS ISSUE the theme is R&D, which is always one of the hardest things to get ones head around. Research and Development plays a central role in the innovation process. It’s basically an investment in potential future capabilities which is transformed into new products, processes, and services. But as Lee Funnell points out in his article this month, the culture and focus of value-add is a fundamental ingredient for success.

In the industry and technology sectors such as our data centre “industry” R&D is a key factor in developing new competitive advantages. Examples of this potential are shown in this issue as a result of PEDCA in the form of new capabilities such as DCA Certifications and new training. The EURECA project as well has the potential to have a hugely positive impact for our sector.

However the prominence of data centres is still relatively low when compared to other industry sectors, but as the importance of the data centre sector to society increases, R&D has to become more prevalent and relevant.Many areas of our industry need further development but as Dr Bashroush points out in his summary of the PEDCA project, an effective framework is needed to bring all the pieces of the jigsaw together.This means improving the co-ordination, engagement and cohesion in both shaping the agenda, policies and representing this aspect to stakeholders must be a key function of the DCA and its members.

How do you buy an energy efficient data centre?

With this month’s theme of research and development and with the EURECA project in mind, perhaps this is one of the industry’s biggest research questions. By Frank Verhagen, CEO at Certios.

WHEN YOU VISIT A DATA CENTRE trade show one is dazzled by the myriad of new products and services proclaiming to save energy and the planet, but still no-one seems to be able to answer the question.

So let’s look at what you can do to answer this question; perhaps the EU Code of Conduct can help? This has the backing of the industry’s leading experts and specialists, it contains the technical guidance and best practices as well as the strategic framework to implement a very well run, energy efficient data centre – so the customer simply chooses an EU Code of Conduct Participating data centre right? Well, not really, unfortunately being voluntary, there is no governance in the field and sadly less than 20% of “Participants” actually report annually as they should. So at best they may be one of the 20% or worse, and more likely, one of the 80% who wear the badge but are simply paying lip-service to energy efficiency.

Ah you might say, but what about PUE or Power Utilisation Effectiveness? This must provide a benchmark that can guide the customer? Well yes and no, unfortunately it doesn’t really tell you anything about the efficiency of the data centre, just how effective the infrastructure may be in distributing power to potentially idle racks.

In addition it again has no governance in its reporting and widely misused and abused. So what can the industry provide? Until recently not much, but at least there is now the DCA Certifications scheme that buyers can trust – not only for energy efficiency by independently verifying best practices and KPI’s like PUE, but also and for some, more importantly for availability classification, operational professionalism and physical security.

Buyers can be assured a stringent set of requirements have been met and they can choose a data centre based on integrity of its credentials – it’s an important step forward for the data centre sector and one which DCA members have insisted on. End users and customers are the true beneficiaries though and to be frank they should, and now can, insist on it themselves.

Here at Certios and being based in the Netherlands, energy
efficiency is high on the agenda amongst policy makers and city planners.

Which is a major reason we bought into the concept of the DCA Certifications scheme at an early stage. We’re now working with data centres both large and small in helping them meet the requirements of the scheme, so that’s testament that the scheme is driving the behaviours that DCA members were hoping for.

Factors influencing the efficiency of
perimeter security solutions for data centres

By Peter Jackson, Managing Director of Jacksons Fencing.

WHILE THE INSTALLATION of physical security products such as gates, bollards and fences to protect a vulnerable data centre site is an obvious requirement, for those responsible for the procurement of these comes the burden to correctly specify those products whose design and manufacture have been proven to deliver the security performance, resilience and longevity required for a high risk environment. Equally important is the commitment to the ongoing inspection and maintenance of all physical security devices once they are installed to ensure their continued efficacy.

Identifying the risk
Following a review of Operational Requirements, usually a two-stage process where the requirement for the security of assets is first established before the solutions are identified, a typical security plan is likely to employ the Onion Principle to provide multi-layered protection to an asset. The aim is to work from the outside into the centre, treating each different boundary as a layer, which is hardened to delay the attacker, provide greater protection to the target and give security staff the intelligence they need to implement their response. This methodology of assessing the perimeter security requirements of a site from the outside in should be applied when determining the most appropriate security solutions for a site – it’s far more realistic and effective than working from a printed plan, a web search, a specification sheet or a fixed budget, from a desk in an office rather than from the perspective of someone attempting to get in.

There’s a lot of discussion around the 5 D’s of effective perimeter security – to deter, detect, deny, delay and defend and perhaps these days, a little too much is centred around electronic detection and surveillance technologies within the hot topic of the Internet of Things and connected devices. But perimeter security is also about creating an effective physical barrier, which can deter, delay and deny unauthorised access to the protected area and ultimately defend the asset. A well planned, designed and installed fence and access solution will add real substance to the protection of a perimeter and can also yield significant savings in the overall cost of site security by reducing the requirement for more expensive electronic alternatives or manned guarding.
Specify for success
Once the threats facing the site have been established and the appropriate types of products designed to mitigate risks and provide the desired level of protection are identified, the focus shifts to selecting the right products from the right producers.
For higher risk sites, various tested and approved or certified products appropriate to mitigating different types of threat are available, which all carry the assurance of proven performance; for example LPS 1175 provides a security rating for products (1 = lowest rating, 8 = highest) which is based on the attack time a product is able to withstand given an allowable tool set and maximum work time. Most sensitive sites would employ products, which have been tested and certified under LPS 1175 while for higher risk applications Approved for UK Government Use products may be recommended by CPNI (The Centre for the Protection of National Infrastructure) as part of their advisory remit. Where Hostile Vehicle Mitigation measures to protect against vehicle borne attack is required, both PAS 68 and IWA 14 certified crash rated fences / barriers / gates and bollards provide a range of effective solutions.

Installing for performance
Any physical perimeter security product, regardless of the testing it has undergone to prove its ability to withstand various challenges is only as good as its installation. It can only deliver to its design specification level if it is regularly inspected and maintained. Therefore, buying decisions need to also consider a number of contributing factors, which affect the integrity of these performance critical products.

For example, materials employed in their construction should be both fit for purpose and able to deliver a long service life. Any steel used which is exposed to the elements above or under ground should at the very minimum be hot dip galvanised to BS EN 1461, inside and out, and ideally after basic manufacturing has been carried out. The steel wire used in the production of mesh fence panels should be coated with a zinc-aluminium alloy treatment in preference to standard galvanised for increased life expectancy. If timber is used it is good practice to ensure that it is manufactured from the right part of the right species of timber so that it is capable of accepting the required timber treatment, which will protect it against rot and wood boring insects.

Attention should be paid to fixtures and fittings - tamper evident, single use or integral concealed fixings on fencing and gates are far less vulnerable than nuts, bolts and rivets. Given that physical perimeter security is an essential part of the infrastructure of a data centre site and crucial to its safe operation, selecting products that are supported by a worthwhile quality guarantee makes a great deal of sense; as along with the peace of mind that comes with having immediate recourse in the event of a product failure, it also ensures costs are contained.

Maintaining performance
The final element to good physical perimeter security is the regular and scheduled inspection of the fence line to identify any changes, which might impact on its security integrity. Litter, debris and dirt should be regularly cleared to allow an unhindered view of the condition of the fence and all repairs required, identified and remedied in a Condition Report.

Gates, turnstiles, barriers and blockers equally need to be regularly tested and checked for mechanical wear and tear, to not only maintain operational efficiency but also to ensure the continued safety of those entering / exiting the site. Particular attention should be paid to the safe operation of automated gates and barriers, which are classed as machinery and as such, are covered by legislation requiring the occupier of the facility to ensure that they remain in safe working order.

It might sound obvious but just because a product hasn’t failed does not necessarily mean that it is working properly and unless there is a standard practice in place to routinely inspect, check, test, maintain and repair the physical elements of perimeter security infrastructure there is the risk that this will go unnoticed.

Visit www.jacksons-security.co.uk for more information on perimeter protection and access control solutions for data centres or call 01233 750 393.

Article provided by Peter Jackson, CEO Jacksons Fencing, Group Vice Chairman, Site Security & Access Control Group, Data Centre Alliance.

Data centre sustainability:
The bigger picture
The data centre industry has done much to improve its energy efficiency in recent years but what about other areas of sustainability? By Operational Intelligence.

A MORE HOLISTIC APPROACH considers the total environmental impact of data centres, looking at the total life cycle impact, not just the energy consumption during the in-use phase.

Research by Whitehead, Shah and others has shown that in addition to the energy use, other important areas of data centre environmental impact are equipment impact of IT and electrical and mechanical services and the energy mix, i.e. source energy, summarised in the diagram below. Areas of protection are human health, ecosystem quality, resource depletion and climate change.

Energy use by power and cooling equipment is generally well understood by the industry, which uses the PUE metric to track performance. The key areas to tackle are air management, use of variable speed drives for fans, higher temperature set points, free cooling and improved UPS efficiency.

Hardware efficiency is continuously improving and IT equipment energy use may also be reduced through increased virtualisation and utilisation, although there is also a trend of increased energy use in networks. Any reduction in IT energy has an associated saving for M&E (mechanical and electrical) energy consumption, so perhaps more focus should be given to making changes here where the impact is greater.

The industry has struggled to develop effective metrics for IT efficiency due to the range of outputs which may be classed as useful work. Jon Summers of the University of Leeds has explored the idea of applying Maxwell’s demon paradox [1] which describes the interaction between thermodynamic and information entropy to develop a metric of IT processing efficiency; initial findings have suggested that IT processing efficiency is seven orders of magnitude less efficient than the theoretical ideal.

Current metrics focus on the in-use phase, however studies have shown that the environmental impact embodied in the equipment and energy source can be equally or more important [2]. There is increased awareness of lifecycle thinking (cradle to cradle), encompassing product design, purchase, recycling and supply
chain issues around social sustainability, e.g. use of rare earth
metals mined in China.

Supply chain codes of conduct do exist and ISO 14001 [3] at one organisation covers energy use, packaging and recycling of e-waste. Procurement officers currently have the wrong incentives and are rewarded for making savings on CAPEX; sustainable procurement should focus more on performance and environmental impact.

Some IT equipment refresh rates are very short (three months), however it is difficult to assess whether the efficiency gain of new equipment is outweighed by the embodied impact. It is thought
that Blade type technology, where the chassis is kept but the blades are refreshed (forward compatibility) has a reduced embodied
impact compared to high volume, pizza box style servers. There is a need for simple, easy to use metrics to help stakeholders evaluate choices.

Often renewable energy is purchased because it is the most cost-effective solution, with the added benefit of green branding; cloud providers have more flexibility over location so are better placed to take advantage of this. Near Niagara in the US, hydropower provides clean electricity for $2c/kWh. Some colocation providers purchase renewable energy for economic reasons but don’t publicise it to avoid committing to clean electricity, though there are indications of market demand for more environmentally responsible behaviour.

Purchasing green energy rather than using on-site renewables may be described as a self-imposed carbon tax and is customer driven. The subject is high profile due to Greenpeace’s Clicking Green report [4], for example. Although there is work being done in these areas, there are many complex issues which need to be tackled with innovative thinking addressing people, policy and process and yielding economic solutions; an ROI approach may not be sufficient to help shift towards lifecycle thinking.

Some issues include awareness and education of those in the industry and lobbyists metrics, changing behaviours, breaking silos between different stakeholders (e.g. IT, facilities, enterprises, vendors, regulatory bodies), communicating the vision in business terms and demonstrating a marketing advantage from lifecycle thinking, and creating a circular economy which addresses all aspects of sustainability.
Data centre sustainability is a large subject and there is plenty of scope for research work to address some of the questions and to inform better practice in a relatively immature area. In order to be financially sustainable other client priorities (availability and capacity) also need to be considered: green dollars as well as the environment.
End notes
[1] https://en.wikipedia.org/wiki/Maxwell%27s_demon

[2.] http://www.datacenterdynamics.com/critical-environment/measuring-environmental-impact-using-lifecycle-assessment/87065.fullarticle

[3.] ISO 14001 Environmental Management Systems http://www.iso.org/iso/home/standards/management-standards/iso14000.htm

[4.] Greenpeace Clicking Green report
http://www.greenpeace.org/usa/wp-content/uploads/legacy/Global/usa/planet3/PDFs/2015ClickingClean.pdf

To meet the needs of tomorrow,
we must foster the culture today
Research and development budgets are often the first to get cut during an economic downturn, yet reducing investment in R&D reduces the potential for economic growth. By Lee Funnell CDCDP, CTP, EMEA Technical Manager for Siemon.

THIS IS A REAL CATCH 22 scenario for companies in the information communication technology (ICT) industry where network speeds continue to advance, data centre growth continues to rise, and an abundance of players results in the need to innovate and stay ahead of the competition. Continually developing new and better products makes it more difficult for others to keep up.

But what exactly does it mean to invest in R&D, and is there a formula for determining the right amount of R&D spending? While investment in R&D can undoubtedly be linked to success (just look at big R&D spenders like Intel, Microsoft and Google), spending more on R&D doesn’t always equate to higher revenue. What it really takes to succeed is innovation—and the culture to support it.
The people have it
The late Steve Jobs said, “Innovation has nothing to do with how many R&D dollars you have. It’s about the people you have, how you’re led, and how much you get it.”

At Siemon, we wholeheartedly agree. You can throw all the money you want at R&D, but true innovation is simply not possible without the core principles and leadership to foster collaborative teamwork, creativity, resourcefulness and integrity.

When a company acknowledges that its greatest resource is its people, and when those people are given the opportunities, the tools and the dynamic environment to perform, it is remarkable what can happen.

Fostering effective teamwork, maintaining an entrepreneurial spirit, and inspiring individuals to think critically and encouraging their ideas and viewpoints are key to that end. Moreover, providing them with state-of-the-art tools, technologies and equipment allows their original and sometimes unconventional ideas to come to life, resulting in some of the industry’s most respected innovations.

Knowledge is power
With an investment in teamwork and individuals also comes knowledge, which is one of the key ingredients for innovation. When a company takes time to share information and ideas among cross-functional teams, they benefit from an influx of knowledge that can be used to further nurture ideas and develop innovative solutions.

For example, ICT companies that actively participate in standards bodies such as IEEE, ISO/IEC and CENELEC and in associations such as the Data Centre Alliance (DCA) can stay abreast of standards development, industry trends and best practices. This arms these companies with the knowledge required for early development of solutions to support emerging technologies and enable the industry to adopt them.

Embrace the customer
Another vital asset from which companies can gain the knowledge they need to innovate are the customers themselves. R&D efforts that are exclusively safeguarded and solely dependent upon internal knowledge and capabilities have a limited ability to result in solutions that customers both need and demand.
To successfully innovate and grow, companies need to open up the development process by collaborating with their customers and listening to their specific concerns and ideas.

Often some of the most innovative solutions are born from the request of a single customer and the realization that the need of one most likely equates to the need of many.
To that end, customer-centric companies that also invest in superior service and support, whilst encouraging sharing of information from the field to the R&D lab, are more apt to develop innovations that solve real-life challenges and gain greater success in the industry.

Closing thoughts
Granting no one can deny that continual investment in R&D is a necessity for staying ahead of the game, the actual amount spent is not always an indicator of a company’s ability to innovate. The degree to which company is truly innovative is based more on the effectiveness and value their solutions bring to the industry. Without a corporate culture that fosters innovation, money spent on R&D has little chance of ever being realised again.

PEDCA project has ended, but the work continues….

By Rabih Bashroush, University of East London.

THE PAN EUROPEAN DCA was proposed for funding to the EU Commission way back in 2011, so it is easy to forget how much the industry has changed over that time, but at that time the DCA was just forming and the idea and concept of the proposal was to develop a “Joint Action Plan” that the DCA, the EU Commission and the industry could follow.

This was achieved with consultation with around 1000 people across 34 countries, 14 regional development agencies, 21 trade associations, 8 standards development organisations and 10 government bodies.

The six Joint Actions which were approved by the commission earlier this year are broadly as follows:

JA1 Technology Leadership Platform:
This involves DCA members playing a greater role in setting the agenda for R&D, disseminating results, co-ordinating and supporting innovation via online and offline activities.

JA2 Representation and JA3 Awareness:
These two actions are related and make up one of the core functions of the DCA, and the project identified how this can be improved and developed.

JA4 Skills and JA5 Training and Education:
These two actions are closely related, and set out where gaps exist and how the DCA can address the “skills gap” and help improve professional development.

JA6 Standards, best practices and certifications:
This set out the strategy for the DCA’s role in assisting with standards development, improving the take up of best practices, such as the EU Code of Conduct. It also provides a roadmap for the DCA Data Centre Certifications scheme, which has now advanced to market.

As DCA members will notice, much of the foundation work of the above actions have already begun, this was a key deliverable of the project in that in developing a plan, each action needed to be underpinned with robust evidence, “real world” testing, detail on service delivery, financial sustainability etc. This was a substantial task, and this work is described in over 500 pages and 85,000 words of deliverables that can be downloaded from the DCA member’s portal.

The skills, training and education aspects of JA4 and JA5 were discussed in detail in last month’s issue, however since then a plan has been agreed at our recent members meeting for 2015/16 which is a direct result of the work undertaken under the PEDCA project and is as follows:
 Assist University of East London to establish portable, low cost,
foundation training course (AKA “Data Centre Essentials)
 The target audience is people not yet employed in industry or
recent new hires
 The outcomes of the course cover the essential industry
knowledge needed by all workers irrelevant of their sector role.
 Develop a BTEC Qualification or equivalent with CDP eligibility
 Run training actions as non-commercial activity – e.g. minimum
overhead to cover the required labour and materials. Meet
requirement for affordability of around €400 per head (as identified
in PEDCA project)
 Make available to commercial training organisations and
Universities as a portable package
 Build a resource pool of qualified trainers that able to support the
delivery
 Aim to complete course design and gain relevant accreditations for
delivery by January 2016
 Investigate and report on options for public funding schemes to support training initiatives
 DCA to construct matrix of qualifications and career path mapping
– provide as a “living document” on website.
 Investigate appetite for “DCA Approved” training and what the
“requirements” might look like.

The course is currently being developed and more details will be available during the autumn.