Why high-performance interconnects are vital to meet the HPC/AI needs of simulation workloads
Webinar There's nothing in our world which exists without being connected whether it is neural, nodal, or networked. And when it comes to organizations working in engineering simulation, digital twins and AI/ML applications, the relentless pressure to improve time-to-results can often require the acme in high-performance interconnects.
Technical computing platforms with HPC clusters that support the design process in simulation have become increasingly important in industries like manufacturing that rely on computational fluid dynamics (CFD) and finite element analysis (FEA) simulations for rapid product design.
Join the Register's James Hayes on 18 October 4.30pm BST/11.30am EDT/8.30am PDT in discussion with Matt Jacobs, CCO at Cornelis Networks, and Eric Lequiniou, SVP of Development at Altair. They'll lay out how to get the most out of fabrics optimized for HPC, AI and data analytics, and explore the importance of high performance interconnects in accelerating and scaling workload requirements while ensuring efficiency and maximum productivity.
Since open source has become increasingly popular in simulation applications development, Cornelis Networks has released an enhanced version of its high performance Omni-Path interconnect, Cornelis Omni-Path Express, based on an industry-conformant enterprise-class open source software stack.
It has worked with partner Altair to leverage the open source-ready version of Radioss Finite Element Analysis dynamic simulation code. That integration has delivered a portfolio of Omni-Path Express accelerated host fabric adapters, edge/director class switches, and gateways designed to provide a powerful combination which can meet HPC requirements for data intensive projects across multiple academic, commercial, government and scientific verticals.
Sign up to watch the webinar here and we'll send you a reminder when it's time to log on.
Sponsored by Cornelis Networks/Altair.