How TotalCAE Enables Multi-Cluster and Multi-Cloud Operations with TotalCAE Multi-HPC

Today’s organizations increasingly maintain computing resources across multiple onpremises high-performance computing (HPC) clusters in various locations.

Organizations may also use one or more cloud providers to have access to state-of-the-art hardware capable of processing and storing the massive amount of data created by HPC and artificial intelligence (AI) workloads.

Organizations often adopt a hybrid mix of multiple onpremises clusters and cloud solutions to meet processing needs. The use of multi-cluster and multi-cloud operations has advantages but leads to challenges for users to utilize.

Challenges Faced by HPC Users

As the number of HPC systems grows—spanning on-premises, cloud, and global clusters — users face challenges in efficiently utilizing available enterprise resources including:

Model data portability: Users face a challenge with data portability when trying to move data that is local to their systems. In the Computer Aided Engineering (CAE) realm, many applications have include files. The include files have hard coded network paths and those network paths are not accessible in other regions of the country, across the globe or in the cloud. Some applications have parts from different teams, and users may not know where all the include files are located which makes running the model or workload overwhelming.