Advanced HPC-CI Webinar Series: Using Expanse AI Resource 101

Remote event

The Expanse AI Resource is available under the NAIRR Pilot program. In this talk we will detail the H100 GPU node architecture, provide examples on usage of AI tools and frameworks, and provide information on the storage options. The AI tools examples will use both conda based installs and Singularity containers compatible with the system GPU drivers. Interactive computing on Expanse AI resource using Jupyter and Galyleo will be covered. Storage options including node local NVMe storage, and Ceph filesystem access both via S3 interface and the direct mount will be shown.

Instructor

Mahidhar Tatineni

Director of User Services, SDSC

Mahidhar Tatineni leads the HPC User Services group at SDSC. He has led the support of high-performance computing and data applications software on several NSF and UC funded HPC and AI supercomputers including Cosmos (PI), PNRP, Voyager, Expanse (co-PI), Comet, and Gordon at SDSC. He has worked on many NSF funded optimization and parallelization research projects such as MPI performance tuning frameworks, hybrid programming models, big data middleware, and application performance evaluation using next generation communication mechanisms for emerging HPC systems.