Data centres – blowing hot air on cold hard cash

DATA CENTRES lie at the heart of the modern economy, from the server rooms that power small and medium-sized organisations to the server farms that run cloud computing services hosted by Amazon, Facebook, and Google. And the explosion of digital content, e-commerce, and internet traffic adds to data centres becoming one of the fastest-growing consumers of electricity.

According to the Natural Resources Defense Council, in 2013 US data centres consumed around 91 billion kilowatt-hours of electricity, which compares to the annual output of
34 large coal-fired power plants. By
the year 2020, data centre electricity consumption is projected to increase to roughly 140 billion kilowatt-hours per year, or the equivalent annual output of 50 power plants, costing American businesses
$13 billion annually in electricity bills and emitting nearly 100 million metric tons of carbon pollution per year.

The European Commission’s 2015 Best Practice Guidelines for the EU Code of Conduct on Data Centres provides an education and reference document, as part of the voluntary Code of Conduct, to assist data centre operators in identifying and implementing measures to improve the energy efficiency of their data centres.

A broad group of expert reviewers from operators, vendors, consultants, academics, professional and national bodies have contributed to and reviewed the best practice, which covers the installation, configuration, and maintenance of the mechanical and electrical plant, racks, IT equipment, and operating systems and software.

Whatever the key driver for businesses in the energy efficiency agenda, driven by rising electricity costs, green legislation, and business sustainability, many organisations have a clear focus on the energy-efficient data centre.

Complying with modern data centre best practice for design and operations is a challenge that must be supported through strategic maintenance. Not everyone can start with a blank sheet and construct a purpose-built, state-of-the-art data centre.

This means that, for most people, retrofit or adaptation and maintaining and efficient and effective maintenance regime is a task that they must tackle. If the data centre is expected to deliver a top performance that is highly efficient, cost-effective, sustainable, reliable and available, it is essential to ensure that the systems are running in a way that can support this level of performance.

The efficiency of air conditioning systems reduces significantly over time, despite the extensive and expensive support contracts that are often in place. Data centres rarely monitor the efficiency of these systems in a proactive manner, and the support contracts are generally only called upon if there is a failure or major problem.

By monitoring the power used by cooling systems, as well as the inlet and outlet temperatures and the air-flow rate, adverse changes in the efficiency of systems can be identified and resolved at little or no cost. Existing problems can also be recognised as soon as a monitoring capability is put in place. A data centre that is not strategically maintained is considerably more likely to suffer downtime and other inefficiencies that impact performance of maintenance can yield a cumulative effect that is even more damaging than downtime.

A lack of appropriate maintenance can cause degradation resulting in users taking longer to perform certain tasks, or automated systems dropping in throughput. Cumulatively this can have a more detrimental effect on productivity than downtime. When the system is down, action is taken but, when degradation occurs, it can largely go unnoticed

Maintenance problems, therefore, may negatively affect performance in small ways that, when considered as a whole, cost a business significant amounts of money.

In 2012, Talk Talk asked Asset Handling to install a condition monitoring solution at their data centres to prevent catastrophic failure. A few months later, they requested further support with gaining deeper insight into their assets including performance, cost, reliability, continuity and risk. The company now uses Asset Insight Manager for all critical assets across their data centres.

Real-time condition monitoring allows Talk Talk to make informed decisions on the maintenance, re-engineering and capital investment required to run the data centre to optimum performance. The system’s unique integration ability, through the simple installation of wireless condition monitor sensors, seamlessly links data feeds to a personalised reporting dashboard.

This removes complex installation processes, and enables simple system expansion,
ease of programming, and end-to-end
asset management. This has enabled Talk Talk to:
 Make informed decisions based
on insights into every aspect of asset
performance using live monitoring
 Visualize the whole operating context of
their data centres
 Manage asset power utilisation
intelligently
 Access meaningful data on asset
performance to run data centres at
optimum performance
 Access real-time quantified asset readings
to increase response times to pre-failure
conditions
 Mitigate risks, spread loads and optimise
space within each rack, cabinet and data
centre
 Reduce costs associated with unplanned
downtime and disruption to business
continuity.