HPC-Cloud-based simulation of drifting snow

The Company

Binkz is an SME which provides state-of-the-art consultancy services involving drifting snow. Roof collapses due to accumulated and drifting snow are responsible for losses of hundreds of millions of Euros as well as bodily injuries and loss of life. This is a problem for all countries in Northern Europe and more generally in the Northern hemisphere. It leads to a costly over-design of buildings, which could be avoided if snow loads could be predicted with sufficient accuracy.  The simulation of drifting snow requires significant compute resources, which can only be provided by a large HPC system, not affordable to an SME. However, the economics of a Cloud-based-HPC solution are very attractive and affordable to Binkz.


The Challenge
The overall challenge was to study the commercial feasibility of a CFD consultation service to civil engineering firms for assessing snow loads on buildings employing snowFoam on the Fortissimo HPC-cloud infrastructure. For the viability of such a consultancy service, it is essential that both the simulation time and the cost of the computation are acceptable within the framework of a typical CFD consultation project.

The Solution
The work in this case study has shown that the simulation of drifting snow using snowFoam is feasible using a Cloud-based HPC system. An analysis can be completed within a few weeks, which fits well with the timescales for the design of buildings. In the solution developed in this case study, the user has access to computing resources, storage and visualization facilities from a desktop environment via a secured webpage in a browser. The required computational resources needed and their costs are appropriate considering those for the overall design of a building.

The Benefits
The simulation of drifting snow requires significant compute resources, which can only be provided by a large HPC system. A typical simulation of drifting snow takes 50,000 CPU hours. This equates to 150 CPUs for 14 days. Furthermore, at a cost of €0.2 per CPU hour, this represents a cost of €10,000. A small consultancy at Binkz would not be able to afford the capital cost of a system containing 150 CPUs, neither could it use a smaller system in-house because the computation time would be much longer than the target two weeks. Even if Binkz were to buy a suitable system, then it would only be used for a fraction of the time and its overall costs would be much greater than the use of a Cloud-based system. Consequently, there is a clear benefit for Binkz in the use of a Cloud-based HPC system.

Participating Organisations
End-user and Code Developer: Binkz
HPC Expert: Vortech
HPC Provider: SURFsara

The Fortissimo Project
Fortissimo is a collaborative project to enable European SMEs to be more competitive globally through the use of simulation services running on a High Performance Computing cloud infrastructure. The project is coordinated by the University of Edinburgh and involves 45 partners including Manufacturing Companies, Application Developers, Domain Experts, IT Solution Providers and HPC Cloud Service Providers from 14 countries. The project is funded by the European Commission within the 7th Framework Programme and is part of the I4MS Initiative.




Other manufacturing

Modelling & Simulation