September 2022 OLCF User Conference Call: Julia for HPC on OLCF Systems
The OLCF hosts monthly User Conference Calls. These calls are your opportunity to speak with center personnel to get the latest updates, express any concerns you may have, etc. No registration is required for this event.
Monthly Topic: Julia for HPC on OLCF Systems
Speakers: William F Godoy, Pedro Valero-Lara, Philip Fackler (ORNL)
In this presentation we will provide a brief introduction to the Julia programming language value proposition for high performance computing (HPC). Julia is a dynamic language designed with scientific computing (Fortran-like) and data science (Python-like) friendly syntax. Julia code builds upon LLVM for performance, prioritizes lightweight interoperability with existing C and Fortran HPC frameworks, while providing a rich environment for data analysis. This unifying and integrated approach targets the scientific computing need of going from idea to performance portability for scientific discovery. In addition, Julia’s design incorporates aspects that are part of third-party ecosystems in other languages, such as: packaging, testing, notebooks for interactive computing, and powerful macro-based instrumentation for productivity, portability and reproducibility. We will present how to get started with Julia from some of our early experiences on OLCF systems: Wombat, Crusher and Summit for running CPU and GPU code with HPC packages available in Julia. We also provide a brief description on how to spin a Julia kernel of OLCF’s JupyterHub, while providing a brief overview of Julia’s own interactive solution via Pluto.jl. Overall, as computing becomes more diverse and heterogeneous, we encourage the OLCF user community to take a hard look at the capabilities and the community around Julia for HPC.