Cloud-based simulation cuts engineers' design costs

Feature

Cloud-based simulation cuts engineers' design costs

One of the big challenges in computer-based design is the amount of processing power required to simulate how designs will perform in the real world.

Electronic simulations of a wind tunnel allow designers to analyse the performance of a component or product without having to build a prototype. Computers allow designers to perform this task, but it has traditionally involved using expensive supercomputers or high-performance computing clusters, which need teams of experts to maintain. For instance, the Renault Formula 1 team has invested £350m in a computational aerodynamics research centre to run race simulations on car designs.

Putting the computational processing power in the cloud reduces this overhead. Southampton start-up dezineforce has developed a hosted engineering design optimisation service. This software as a service (SaaS) offering gives users access to an integrated suite of tools for the analysis of design behaviour and for systematic search.

Collaborative working

As well as design optimisation software, the suite includes high-performance cluster computing and simulation applications for structural, fluid dynamics and mechanism analysis using Microsoft Windows HPC Server 2008. Because the tools are delivered over the internet, they can be accessed from anywhere in the world, by whoever the subscribing company chooses, facilitating collaboration between teams at different locations.

The team behind dezineforce include professors Simon Cox, director of the Microsoft Institute for High Performance Computing at the University of Southampton, and Andy Keane, director of the Rolls-Royce University Technology Centre for Computational Engineering at the University of Southampton. Customers include global design and engineering firm Arup, along with small design consultancy businesses and specialist manufacturers.

The company has shown what SaaS can do for engineering at several events this year. It recently demonstrated a wind farm optimisation simulation with a company called Intelligent Fluid Solutions, which used the dezineforce service to simulate the optimal position for wind turbines on a wind farm. The service allowed Intelligent Fluid Solutions to run the simulation as a cloud service, without the need to invest in a high-performance computing cluster.

Microsoft platform

The dezineforce service uses a Microsoft platform which includes Internet Information Services (IIS) 7.0 and ASP.net, together with Ajax programming. A 3D visualisation tool is implemented as a Windows Presentation Foundation (WPF) browser application, which provides hardware-accelerated, interactive 3D visualisation within a browser, without the need to install software.

Cox says dezineforce is more suited to the way engineering projects are run, compared with other ways of providing high-performance computing. "Engineering is very project-based. There is an intense level of activity then a quiet period," he says.

An in-house high-performance computing cluster will be mostly idle during these quiet periods. As such, Cox says high-performance computing in the cloud is a better option, because users only pay when they need the computational resource.

The company says its IT infrastructure has been built to run, schedule and secure multiple engineering simulations in the cloud.

Cloud computing certainly looks like the next big wave that is hitting the IT industry. If dezineforce is successful in convincing businesses both large and small to run computational simulations in the cloud, it will prove that cloud computing can be used for some of the world's most complex computational tasks.


Email Alerts

Register now to receive ComputerWeekly.com IT-related news, guides and more, delivered to your inbox.
By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy

This was first published in July 2009

 

COMMENTS powered by Disqus  //  Commenting policy