hpcDIRECT scales new heights with SATAVIA

HPC Insights


As I take a short break from the exhibition floor here at another excellent ISC18 in sunny Frankfurt, I’m delighted to announce SATAVIA as the latest customer to join our bare metal high performance computing (HPC) platform – hpcDIRECT.


Founded by Dr. Adam Durant in 2013 and based in Cambridge - one of the UK’s central hubs of science and technology innovation - Satavia has developed an environmental data intelligence platform which enables airlines to reduce the financial impact of unscheduled engine and airframe maintenance caused by ice, dust and corrosive air pollution – something I am sure all the delegates in Frankfurt will be pleased to hear about when we all board flights and fly home from here tomorrow.

Like me, you are probably continually bombarded with companies of all shapes and sizes telling you they’re being innovative, or their latest product is the most groundbreaking one on the market since, well, their last innovation. It all gets a bit tiresome when you realise there are only so many ways to crack a nut and we don’t need another ten. So, against that backdrop it’s refreshing to come across a company like Satavia who are genuinely doing something pioneering and smart, and breathing new ideas and techniques into a long-established and very old school industry.

Satavia’s data intelligence platform is the only solution that combines and validates multiple environmental, weather and aircraft datasets. Through this compute-intensive research Satavia can analyse years of historic data that will supply aviation and aerospace organisations across the globe with mission-critical information and go a long way to helping ensure these industries operate smarter, cleaner and safer. Their team of data scientists and machine learning experts generate exposure analytics for individual aircraft, integrating machine learning to offer automated risk-based decision-making capabilities – another timely example as we approach ISC18’s ‘Machine Learning day' tomorrow, of the increasing convergence between HPC and AI.

Supported by our excellent in-house HPC specialists at Verne Global (who have been helping Satavia’s team across the full lifecycle of their initial testing stage – a service we offer all companies interested in trialing hpcDIRECT), Satavia will gain flexible and scalable access to 10,000 bare metal HPC cores utilising latest specification Dell EMC and Intel technology. By using hpcDIRECT their team intends to map out the prevalence of high altitude ice clouds which pose a hazard to aircraft and can cause engines to flame out during flight. International aviation and aerospace organisations will rely on the information gleaned from this research to develop and implement innovative solutions to boost efficiency, improve safety, as well as to reduce costs and aircraft emissions.

Another pleasing thing about working with Satavia was their clear determination from day one to ensure their high performance computing was deployed as sustainably as possible. This made Iceland the obvious choice due to the fact Iceland remains the only developed country that generates all of their power from 100% sustainable sources (geothermal, hydro and wind). Plus, through Iceland’s temperate North Atlantic climate and Verne Global’s optimised campus infrastructure our hpcDIRECT platform is chilled by 100% cool, Icelandic ambient air. While Iceland isn’t great for sunbathing, it is perfect for keeping power-intensive, industrial scale HPC hardware at optimal operating conditions without compromising the environment.

Given the global data center industry now produces more carbon emissions than the entire global aviation industry, the urgent need for more companies to follow Satavia’s lead is clearly paramount. Not that anybody needs one, but that’s another good reason to come out to the fantastic country that is Iceland and see how we’re ensuring big data is green data.

Since hpcDIRECT was launched at the end of last year, our first six months have provided our team with unique opportunities to work hand-in-hand with our client’s HPC engineers on the development of their operations and deployment of their applications. I think it’s as close to a tailored, ‘cut to fit’ service as you could get and the supplier-buyer interaction on hpcDIRECT is thoroughly exciting and rewarding to be part of. Common to all of the companies utilising hpcDIRECT is a requirement to utilise a truly industrial scale, TrueHPC environment for their workloads that can flex and scale when needed, and that has sustainability as a key parameter within their compute portfolio.

Back to ISC18 - If you haven’t yet come to see us at booth F-912, please do, and I look forward to meeting you. By undertaking a short five question survey you have the opportunity to win a fantastic trip for two to Iceland. Rest assured, you can leave the sun-tan lotion at home for that one....


Written by Spencer Lamb

See Spencer Lamb's blog

Spencer is Verne Global's Director of Research and head's up our high performance computing work with European research and scientific organisations. He is also a member of the European Technology Platform for High Performance Computing (ETP4HPC).

Related blogs

Cloudpoint: A blueprint for high performance computing

As part of the hpcDIRECT team, based in the London offices of Verne Global, we continue to develop and refine our "high performance computing as a service" (HPCaaS) product that was launched in December last year.

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


SC18 - My Schedule Highlights

SC18 gets underway in Dallas, Texas, today, kicking off a week of panels, workshops and tutorials. This is the 30th year of the conference, which is properly known as the International Conference for High Performance Computing, Networking, Storage and Analysis. It's an exhausting schedule to navigate and even experienced attendees will have to do some diligent research to pick out the highlights. Here are mine...

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.