SPONSORED CONTENT HIGH PERFORMANCE COMPUTING
international efforts on emerging high performance computing architectures. Making IFS code open source should encourage work on the IFS by computational science experts in academia and vendors. The IFS GitHub space contains code
released primarily to support pre-existing collaborations. While available to anyone, code in the IFS space is not generally supported. It currently contains the following elements: ecRad, ecTrans, FIAT, CLOUDSC, CLOUDSC2 and ecBundle. These tools help to predict spectral storms and cloud microphysics, among other features.
Weather in the clouds Today, the world’s leading weather- forecasting organisations usually process weather data four times a day, which, depending on the speed of execution, can take around two hours at a time. This can be very expensive when using the cloud, resulting in companies only using cloud capacity for eight out of 24 hours, leaving a lot of downtime still to be paid for. Tomor row .io began its business providing
weather intelligence. Their weather and climate security platform help organisations – from Ford, Porsche and Uber to Google and Fox Sports – to predict the business impacts of weather and solve their climate
security challenges. By producing business insights and action plans calibrated to a minute-by-minute forecast, they can automate decision-making at scale, helping countries, businesses and individuals better manage their weather-related challenges. To achieve forecasting and analysis at speed and scale, Tomor row .io needed to harness the power of high performance computing (HPC) to run its numerical weather prediction models. As a start-up they used the public cloud, which gave them instant access together with a degree of flexibility. But as the business expanded and grew its operational and research offering, their HPC costs became increasingly expensive and more complicated to maintain; they found they were paying a premium for a service that did not always deliver what they needed, and though they considered other public cloud options, the potential for making savings proved very limited. Getting the power they needed to analyse data for the industry standard four times a day was proving an issue, with Tomor row .io needing to increase data processing time so that it became a near-constant requirement. The team at Tomor row .io approached
atNorth to see if they could power their next phase of research and development
(R&D) more cost-effectively. They wanted to change the rules of weather forecasting, analysing data constantly, in real time, to provide on-the-spot calculations that would allow their high-end clients to make business-critical decisions with confidence. As a meteorological organisation with a
strong focus on environmental, social and governance (ESG) issues, Tomor row .io was interested in atNorth’s cloud capabilities, but also in its sustainable, 100 per cent renewable energy-powered data centres, and their commitment to circular economy principles. Keenly aware of its carbon footprint and ready to minimise it, Tomor row .io was looking for a solution that would not only save money for clients, but would also be good for the planet. Luke Peffers, chief weather officer at
Tomor row .io, commented: ‘We are a growing company with massive scale as our goal. One of the consequences of this is that we run large jobs for our customers that can eat away at our margins. atNorth offers us an excellent solution at a discount so we can scale as an organisation and pass the cost savings to our customers. Having a lower price and better performance from a sustainable source is a win-win situation. We love the cost savings as well as the sustainability, and so do our clients.’
New White Paper now online
VIEW FOR FREE*
Flexible, scalable and sustainable: The new power behind tomorrow’s
weather predictions The world’s leading weather-forecasting companies usually process weather data four times a day, which, depending on the speed of execution, can take around two hours at a time. This can be very expensive when using the cloud – resulting in companies only using cloud capacity for eight out of 24 hours, leaving a lot of downtime still to be paid for.
SCIENTIFIC COMPUTING WORLD
www.scientific-computing.com/white-papers
www.scientific-computing.com | @scwmagazine Summer 2022 Scientific Computing World 15
elRoce/
Shutterstock.com
*Registration required
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38
