Clean air takes fresh thinking

HPC Insights


You might think that the issue of air purity in data centers is done and dusted, so to speak. We all know how to handle electronics, and the dust and pollutants in air are well understood. However it turns out that air contamination is a live issue and attempts to clean up other parts of the data center environment could have unintended consequences on the atmosphere inside the data center.

Firstly, regulations have removed lead from the solder that is used to attach components to printed circuit boards (PCBs). The European Union’s Restriction of Hazardous Substances (RoHS) Directive of 2006 banned lead in solder, in order to keep it out of electronic waste, and this effectively ruled lead out for the global electronics industry - though some people questioned the logic, as in fact very little lead escapes from waste into drinking water.

What’s that got to do with air purity though? Without lead, solder is less robust. It’s more susceptible to corrosion caused by pollution, and it also grows tin whiskers, which can break off become airborne and cause short circuits. Without lead, our electronics became more susceptible to air pollution - and started to produce their own.

At the same time, to make data centers use less energy and reduce their carbon footprint, designers started using outside air to cool the facility. It’s more efficient than cooling air down inside the building - but it adds to the risk of outside contaminants getting in.

“Contaminants can be introduced during infrastructure upgrades, or by simply opening doors while entering or exiting a white space,” say Paul Finch of Kay Data and Chris Muller of Purafil in a recent DCD article. “These contaminants affect electronic equipment, corroding components and reducing capabilities to failure point, which can result in costly data center outages.”

And finally, the location of data centers is changing. Edge computing means putting resources close to the end users and devices that need them - and that tends to mean digital infrastructure being placed in more urban areas, where there is a risk of particulates from diesel engines along with sulphates and other impurities.

Even some remote areas have special considerations: Iceland for example generally has some of the purest air around, but it has infrequent volcanic eruptions. In 2010, one of these produced enough particulates to ground flights across Europe (though ironically not flights from Keflavik International airport, or the main data centers locations which are 250km away and upwind of any volcanoes).

There are recommended procedures to deal with all these risks, like any other risks - and these boil down to understanding what is going on, planning, monitoring, and having a response ready.

The first thing is planning: outside air can be run through a heat exchanger for indirect cooling, and filters should be put where there is a risk of contamination.

Back in Iceland for instance, Verne Global’s campus is upwind has a wall of air filters (shown in the photo above) that can take out volcanic ash particles within air. As rare as an eruption is, if the levels ever got too high, the filters can be shut, and the facility will run on recycled air. Urban data centers in places like London, New York and Paris have to make similar plans for constant road traffic pollution.

Monitoring is straightforward: sensors can measure contaminations, and these should be networked wirelessly, to give a real-time picture of air in the data center, as well as archived reports of the quality, to give advanced warning of any issues.

All data centers are designed for a level of reliability: keeping the air clean is just one part of that.


Written by Peter Judge (Guest)

See Peter Judge (Guest)'s blog

Peter Judge is the Global Editor at Datacenter Dynamics. His main interests are networking, security, mobility and cloud. You can follow Peter at: @judgecorp

Related blogs

G-Cloud 10 makes accessing high performance computing easier then ever...

As the Director of Research at Verne Global I spend a lot of my time working with our colleagues and partners within the UK’s publicly funded universities and research and science community. I’m privileged to get to see some of the truly innovative and inspiring research that is taking place, using high performance computing (HPC) and further encouraged with how Verne Global is helping them do this. This is why I was delighted to see Verne Global’s participation in the G-Cloud 10 (G10) framework confirmed last week and indeed strengthened for 2018/19 – enabling more public sector bodies to enjoy the benefits of our on-demand true hpcDIRECT platform.S

Read more


Explainable AI

SC18 here in Dallas is proving once again to be a fascinating melting pot of HPC insights and observations, and it's intriguing to see the continuing convergence of AI into the supercomputing ecosystem. Along these lines I started to think about the movement towards 'Explainable AI'. Being able to explain and understand how models work when making predictions about the real world is a fundamental tenet of science. Whether solving equations in a dynamic system for precise answers or using statistical analysis to examine a distribution of events, the results sought from these methods are intended to increase our clarity and knowledge of how the world works.

Read more


DevOps. Stacking up to be a Common Theme at ISC18

Yesterday was a good day to be in Frankfurt. All of the majors in the supercomputing universe descended upon the Messe Frankfurt to begin ISC18 with a series of training seminars. For my morning session, I chose Getting Started with Containers on HPC through Singularity put on by the team at Sylabs, Inc. I have been tracking the progress on Singularity in the HPC community since before Sylabs was founded by CEO Greg Kurtzer in an effort to bring the technology of root secure containers into the realm of enterprise supported software. I was excited to hear about the progress that Sylabs has made and to see where the future of containers lies for the broader HPC community. If I was forced to sum the tutorial into a single portmanteau, it would be DevOps. After this session, it is clear to me that the world of DevOps that has been created in the cloud native universe is on a collision course with HPC. And the future of science says that it can’t happen soon enough.

Read more

We use cookies to ensure we give you the best experience on our website, to analyse our website traffic, and to understand where our visitors are coming from. By browsing our website, you consent to our use of cookies and other tracking technologies. Read our Privacy Policy for more information.