https://julialang.org

Julia is a modern open-source language and unified ecosystem for science. Powered by the widely vendor-adopted LLVM compiler infrastructure, Julia attempts to close the gaps in high-productivity and high-performance, in which the combination of Python with traditional compiled languages C, C++ and Fortran and their ecosystems, Python+X, can be a costly development and maintenance investment for scientific projects. Julia offers out-of-the-box abstraction for multidimensional arrays, access to CPU and GPU computing, a rich mathematical standard library, and a high-level interface for data analysis, visualization and interactive computing (the Julia REPL, Jupyter and Pluto.jl notebooks). In addition, several aspects that traditionally belong to a language ecosystem are core aspects of Julia: packaging, reproducible environments, powerful macros metaprogramming for code instrumentation (profiling, testing, etc.) and interoperability with existing Fortran, C and Python codes. Thus, Julia provides an interesting investment when trying to find a balance between development costs and scientific discovery.

Aims and Scope

After a successful first edition in 2022 with 101 registrations, lightning talks and discussion sessions, the Second Julia for ORNL Science Workshop, JuFOS 2024, invites the lab's scientific community to share current challenges and opportunities for their computing needs. We aim to provide a venue to discuss the potential of Julia as a language for science, and its ecosystem, current uses, and community status updates. Our goal is to provide a venue to exchange research and development ideas about current state-of-the-art development of scientific codes at ORNL and potential collaborations and investments around a relatively new language designed for science. We welcome applications from the wide range of experimental, observational, high-performance computing (HPC) projects at ORNL.

Call for Lightning Talks

We invite ORNL participants to present a 3 to 5 minute lightning talk that aligns with the scope of the workshop (option provided in the registration form). Talks are meant to be informative, (e.g. requirements for high-level languages) and not restricted to the topics below.

Topics:

Registration

While this is a hybrid event, we encourage in-person participation for the morning, working lunch, lightning talks and panel sessions to learn more about the language and its community. We invite attendees to share their work regarding with the lab community by submitting a 3-5 minute lightning talk. Lunch, snacks and coffee will be provided for in-person attendees. Virtual attendees will receive a Zoom link to join the event.

Registration Link: Please register for virtual participation by June 17th 2024 using this link.

The registration form includes the option to propose a lightning talk.

Virtual Tutorial Events

These virtual events are offered the week of the workshop. Separate registration is required, please visit the events site. Registration for the workshop DOES NOT include registration for the virtual tutorials.

Agenda

Thursday, June 20th 2024 at the JICS Auditorium, Building 5100

Time Session Presenter
Morning    
10:00am-10:30am Welcome, updates, brief intro to Julia William Godoy
10:30am-11:00am LLMs and easy performance portability with JACC.jl in Julia Pedro Valero-Lara
11:00am-11:45am Invited Talk: “Sunny.jl: lessons learned from a DOE scientific application” Kipton Barros, LANL
11:45am-12:10pm Break  
Working Lunch    
12:10pm-12:50pm Invited Talk: “Heterogeneous computing with the Julia language: from A64FX to the IPU” Mosè Giordano, UCL/ARC, UK
12:50pm-1:00pm Break  
Afternoon    
1:00pm-2:00pm Lightning Talks/Panel Session 1: Methods and Infrastructure Session Chair: Het Mankad
  JuliaIRIS: Julia with IRIS Heterogeneous Runtime System Narasinga Rao Miniskar
  Performance Portability with Julia: My Internship Experience Claire Winogrodzki
  Julia Implementation of a Fast Heuristic Solver for Binary Optimization Ryan Bennink
  Evaluation of the convolution operator in a Julia machine learning library Zheming Jin
  Novel iterative methods for multiphysics problems Nicolás Barnafi
  Pkg - Julia’s Integrated User-Friendly Package Manager Kelly Tang
  The Experimental Computing Laboratory (ExCL) Steven Moulton
  Panel Discussion  
2:00pm-2:15pm Break  
2:15pm-3:45pm Lightning Talks/Panel Session 2: Applications Session Chair: Het Mankad
  Harnessing the Power of Sets in Julia Youngsung Kim
  Integrating high-throughput ODEs and Molecular Language Models in Julia and Pytorch Zachary Fox
  Node-level regression task to predict vaccine hesitancy from graph learning Sifat Afroj Moon
  Remotely-sensed change detection with Julia David Hughes
  AutoCSM - Automating Supercomputer Cooling System Model Generation Scott Greenwood
  Must.jl Multiple Scattering Theory for First Principles Calculations of Materials Markus Eisenbach
  Julia for HPC- Porting to AMD GPU devices for Climate Applications Antigoni Georgiadou
  Integrating Julia and OLCF HPC infrastructure to enhance SNS data processing capabilities Steven E. Hahn
  Panel Discussion  
3:45pm-4:00pm Closing remarks, adjourn Pedro Valero-Lara, William Godoy

Organizers

Sponsors