In this track dedicated to engineering computing, Craig Hamill, Former Research, Innovation, and Technology (RIT) team lead at UL Research Institutes, discusses the use of a hybrid setup--a combination of Microsoft Azure and on-premise clusters--to conduct safety science analysis, from modeling fire scenarios to performing millions of material derivations for prototype development.
The setup at UL Research reduced queue wait times by 60%, yielding answers to research questions in days instead of weeks, Hamill recalled. As he sees it, increased model complexity is driving on-premise simulation queues to become bottlenecks. Relief comes in the form of cloud burst capacity that augments on-premise setup.
Hamill also discusses practical, logical paths from a workstation-based workflow to the cloud, ways to estimate the cost of simulation runs to stay under budget, the often-overlooked costs in maintaining and operating a data center, and the impact of AI on HPC workloads.
Register here to submit your questions and watch the session.

Kenneth Wong is Digital Engineering's resident blogger and senior editor. Email him at [email protected] or share your thoughts or suggestions at digitaleng.news/facebook.
Follow DE
Join over 90,000 engineering professionals who get fresh engineering news as soon as it is published.