Workshops Archive

A Distributed HPC/AI Services Ecosystem for the Genesis Mission Platform


Workshop: High Performance Python for Science at Scale

Authors: Aymen Alsaadi and Andre Merzky (Rutgers University); Andrew Shao (HPE); and Shantenu Jha (Rutgers University, Princeton Plasma Physics Lab)

Abstract: Modern HPC and AI applications require capabilities deployed, composed, and accessed on demand. However, reusable services must coordinate heterogeneous resources, lifecycles, and low-latency data exchange. We present a distributed service ecosystem for the American Science Cloud (AmSC) of the Genesis Mission, providing mechanisms for deploying, composing, and accessing HPC/AI capabilities. Our contribution is a service-building ecosystem for HPC/AI workflows that separates service management, workflow execution, and data exchange, enabling capabilities to be composed without application-specific integration. We demonstrate the approach through an asynchronous OpenFOAM optimization workflow, showing how HPC/AI capabilities become reusable services across AmSC resources.


Back to High Performance Python for Science at Scale Archive Listing

Back to Full Workshop Archive Listing