Today Altair announced that eleven international customers participated in the company’s recent HPC Cloud Challenge. The contest was set up to demonstrate the benefits of leveraging the cloud for large-scale design exploration in the area of computer-aided engineering. Organizations of all sizes from manufacturing and academic fields participated in the Challenge, utilizing Altair technologies in structural, CFD and design studies, and expressed great satisfaction with the program overall.
Altair is very excited to see the results of the HPC challenge,” said Robert Walsh, Altair’s Vice President Business Development for Cloud Solutions. “Projects like this help validate this new HPC paradigm for everyone. This type of solution demonstrates the true value and power cloud based HPC for CAE. With the combined solutions of Altair Engineering, AWS and Intel now cutting-edge tools can truly be democratized by reducing barriers for highly complex engineering tasks, bringing performance, flexibility, scalability and affordability to CAE, triggering simulation-driven innovation.”
Qualified participants were provided access to Altair’s software, computing resources from AWS, and product support for the duration of the program. Participants could choose from among six different configurations, starting with 32 cores and going up to 128 cores or more on the Intel Xeon processor E5 product family, and Altair’s HyperWorks Unlimited (HWUL) Virtual Appliance, a software solution for high-performance computing on the cloud. All participants accessed the solutions free of charge for seven or more days to complete the engineering tasks of each project. The free use of Altair’s HWUL Virtual Appliance was supplemented through AWS credits. All together the customers submitted a total of 43,520 jobs, running 10,405 wall clock hours. The HyperWorks solvers most applied in the challenge were RADIOSS, OptiStruct and HyperStudy.
“We tried HyperWorks Unlimited Virtual Appliance for the first time via the HPC Challenge and I have to say we were extremely satisfied – it’s pretty simple, robust, fast and reliable,” said Fernando Mitsuyassu from Maxion Wheels. “We usually handle our simulation tasks on local workstations and this was much faster. The reduction of processing time we reached was around 50 percent and in some cases even up to 80 percent.”
For us the HPC Challenge was a very positive experience,” said Dr. Jörg Büttner at ZF TRW. “We used the HPC Challenge for a project for which we applied the RADIOSS solver with 32 nodes, explicit. A solution such as HWUL Virtual Appliance enables parallel work, is very easy to use and provides a fast up- and download of data. With more resources available, and by using cloud and local resources in parallel, we received more results faster.”