The Oak Ridge Leadership Computing Facility (OLCF), in conjunction with the Oak Ridge National Laboratory Computer Science and Mathematics Division (CSMD), will host Julia for HPC, a hands-on tutorial introducing participants to using Julia for high performance computing (HPC) applications. This training is open to NERSC users.
This session will explore advanced topics relevant to HPC, including GPU programming and performance-focused Julia packages.
Who should attend
This session is open to all who want to learn more about using Julia for HPC and scientific computing. Users of OLCF, SNS, HFIR, NERSC, and ALCF resources are especially encouraged to attend.
Presenters
William Godoy, Philip Fackler, and Pedro Valero-Lara (ORNL)
Training modules
- Brief introduction to Julia and the HPC ecosystem
- JACC.jl for performance portable science codes: XSBench.jl, miniBUDE, Hartree-Fock, LULESH
- Developing and running a Julia HPC diffusion-reaction app: Gray-ScottSimulation on CPU/GPU using JACC.jl
- Communication via MPI.jl
- Parallel I/O with ADIOS2.jl
- Performance tips
- Package interactions and configuration with a HPC system
- Interactive computing with Jupyter and Pluto.jl notebooks
Compute resources
Participants will need access to HPC systems for this session. Please apply for access to OLCF’s Odo training system project TRN036. Odo’s architecture is similar to that of Frontier.
OLCF account applications must be received by June 19, 2025.
Training Materials
Julia v1.11 release candidate
8 days ago
HPC Gray-Scott example
8 days ago
Julia HPC miniapp using parallel models (MPI.jl, CUDA.jl, AMDGPU.jl, ADIOS2.jl) and Jupyter/Pluto.jl notebooks
HPC Gray-Scott example
8 days ago
Julia for High-performance Computing Tutorial
JACC
8 days ago
CPU/GPU parallel performance portable layer in Julia via functions as arguments
JACC
8 days ago
A collection of applications that make use of JACC.jl as examples/benchmarks
Julia’s Value Proposition for Better Scientific Software
8 days ago
Recommended reading