Building a Smarter Planet. A Smarter Planet Blog.

By Harry van Dorenmalen
Chairman, IBM Europe

Recently, IBM detailed two new computers that will help change the way the world, literally, works.

The first, Sequoia, is the world’s most powerful supercomputer, capable of calculating in one hour what would otherwise take 6.7 billion people using hand calculators 320 years to complete if they worked non-stop. It is installed at the National Nuclear Security Administration (NNSA)’s Lawrence Livermore National Laboratory in California.

The second is the first commercial machine, cooled by hot water, built for the Leibniz Supercomputing Centre in Germany. It will be used by scientists across Europe to drive a wide range of research ­­− from simulating the blood flow behind an artificial heart valve, to devising quieter aeroplanes.

What’s impressive about these machines is not just their massive processing power alone, but  they are remarkably energy efficient, too.

For example, while Sequoia is more than one and a half times faster than its nearest competitor, it is almost two and a half times more energy efficient. Meanwhile, the water-cooled Leibniz machine consumes 40 percent less energy than traditional air-cooled versions. Since water is able to remove heat 4,000 times more efficiently than air, hot water is an innovative means to more efficiently cool computers than the conventional approach based on chilled water. It also allows energy to be captured from the system and reused to heat the buildings during winter. These advances afford potential energy savings of around 1 million euros per year.

Not all large-scale supercomputers need to be quite as powerful as these two. However, across Europe hundreds of machines are used to run essential daily tasks in the numerous data centers that power many of the government and commercial systems we take for granted.

As a result, information and communication technology already consumes 8 to 10 percent of the EU’s electricity and by 2020 that figure may double. Data centers in Western Europe alone consume up to 100 billion kilowatt hours a year. That’s more or less the total electricity consumption of the Netherlands.

That’s why the European Commission has created a Code of Conduct to encourage all data center operators to reduce energy consumption in a cost-effective manner. The guidance provides a set of best practices.

In May 2012, the European Union Code of Conduct for Datacenters pronounced IBM the Corporate Participant winner for 2012. Dr. Paolo Bertoldi, Principal Administrator at the European Commission Joint Research Centre in charge of research activities on energy efficiency policy, praised IBM for its systematic approach and for achieving results in a large number of data centers.

It’s the latest in a line of environmental awards since IBM announced its first corporate policy on environmental affairs in 1971.

What does IBM do to maintain its energy efficiency leadership in its data centers? Older hardware equipment is continually upgraded while industry-leading server consolidation and virtualization techniques are also employed to help ensure computers run at optimal capacity.

Analytics plays a crucial role too: Mobile Measurement Technology (MMT), developed by IBM Research, instruments the data centers with thousands of sensors to record and analyze temperatures and air flow to detect hot and cold spots. By providing energy flow insight in real-time, MMT provides the intelligence to enable more efficient cooling.

In an increasingly instrumented and interconnected world, the amount of data the world is producing is growing exponentially; we reckon there will be around 44 times as much data and content over the next decade as there is today.

Gaining insight and intelligence from all that data will largely be the job of the world’s supercomputers and that will take a great deal of processing and energy power. We’ll continue to work tirelessly on increasing the first while systematically reducing the second.

Technorati Tags: IBM, Lawrence Livermore National Laboratory, Leibniz Supercomputing Centre, Sequoia