Join Oak Ridge National Laboratory scientists William F Godoy and Philip W Fackler for a live webinar exploring how JACC.jl enables vendor-neutral parallel computing in Julia across CPUs and GPUs.
High-performance computing development often requires engineers and researchers to manage multiple programming models and vendor-specific GPU frameworks. JACC.jl simplifies this process by enabling portable Julia code that runs across CPUs and GPUs from different vendors, all from a single source code base.
In this webinar, Oak Ridge National Laboratory experts will introduce JACC.jl and demonstrate how Julia users can build vendor-neutral parallel applications while maintaining performance and developer productivity.
Through live examples running directly on laptops, attendees will see how JACC.jl helps accelerate HPC adoption using modern Julia tooling.
Presenters:
William F Godoy, Senior Computer Scientist
Philip W Fackler, Research Software Engineer
Topic: JACC.jl: Julia for ACCelerators
What You’ll Learn
How JACC.jl enables vendor-neutral CPU/GPU computing in Julia
Writing portable parallel Julia code from a single codebase
Deploying Julia applications across multiple accelerator vendors
Best practices for HPC development using Julia
Real-world use cases and live demonstrations with JACC.jl
About JACC.jl
JACC.jl (Julia for ACCelerators) is a registered Julia package designed to simplify portable CPU/GPU parallel programming. It enables developers to deploy a single source code across CPUs and GPUs from different hardware vendors with minimal effort.
