More from the team of Wolfgang Gentzsch and Burak Yenier on their experiment designed to explore cloud computing and CAE, and the potential of HPC-as-a-Service. (See related Digital Manufacturing Report story here.)
In a feature that appeared today in our sister publication, HPC in the Cloud, Gentzsch and Yenier take a more detailed look at the current situation in manufacturing. They cite research from NCMS and other sources that reveal the paltry use of HPC in manufacturing companies (only 15%) despite the need of many of these small to medium sized firms for the compute power required to tackle problems that their existing desktop systems can’t handle.
They present a CFD use case scenario to contrast the results of running an iterative CAD/CAE process to modify a product’s geometry using high-end workstations vs. running the same job in the cloud. Amazon’s EC2 Compute Cluster Instances are used in the example. No surprise that the cloud approach wins hands down.
Comment Gentzsch and Yenier, “This simple example demonstrates that a much bigger job, which was impossible to execute on a workstation, can be run on Cloud resources in a short time and at a reasonable price.”
But they also add this caveat: “…we don't mean to suggest replacing the engineer's workstation and ISV's software license on that workstation with Cloud resources. These tools will continue to be invaluable for the engineer and his R&D projects. We view Cloud computing and on-demand software licenses as an additional, complementary benefit for the engineer's daily work.”
They also reiterate their call for participation in the CAE experiment and provide links to sign up and obtain more information.
You can read the full story at HPC in the Cloud by clicking on the link below.
Full story at HPC in the Cloud