Democratizing HPC Access and Use with Knowledge Graphs
Pouya Kousha, Vivekananda Sathu, Matthew Lieber, Hari Subramoni, Dhabaleswar K. DK Panda · 2023
The field of High-Performance Computing (HPC) is undergoing rapid evolution, with an expanding and diverse user base harnessing its unparalleled computational capabilities. As the range of HPC applications grows, newcomers to the field are faced with the daunting task of optimizing their applications for efficient execution on HPC systems. Traditional documentation, often spanning dozens of pages, is cumbersome for finding answers and ill-suited for integration with emerging conversational AI-powered user interfaces like chatbots. Addressing this challenge, we propose a novel HPC ontology crafted to encapsulate HPC runtime relations in a scalable fashion. Our proposed ontology not only facilitates the transfer and querying of this knowledge but also serves as a foundational pillar for our AI-powered Speech Assistant Interface (SAI)[13]. This ensures reproducibility, reliability, and optimal performance when executing tasks. In this paper, we elucidate the relationships and properties underpinning our ontology and showcase how users can interact with knowledge graphs based on our proposed ontology to derive insights.