2021 ALCF Simulation, Data, and Learning Workshop

Workshop
2021 ALCF Simulation, Data, and Learning Workshop

The ALCF's annual Simulation, Data, and Learning Workshop will be held virtually October 5-7, 2021. The interactive workshop is aimed at researchers with near-term goals of applying for a major allocation award.

The ALCF's Simulation, Data, and Learning Workshop is designed to help researchers improve the performance and productivity of simulation, data science, and machine learning applications on ALCF systems. Workshop participants will have the opportunity to:

  • Work directly with ALCF staff experts during dedicated hands-on sessions
  • Learn how to use available tools and frameworks to improve productivity
  • Test and debug codes with exclusive system reservations on ALCF computing resources
  • Get assistance with Director's Discretionary projects to help prepare for a major allocation award
  • Improve application performance for current ALCF projects
  • Plan ahead for 2022-2023 allocation proposal submissions
     

REGISTRATION DEADLINES

  • Foreign Nationals: September 13, 2021
  • U.S. Citizens: September 20, 2021

Note: Registrants will be reviewed for experience level and will be asked to provide goals for attending.

If you have any questions, please contact us at sdl-workshop@alcf.anl.gov

Agenda

Day 1: Tuesday,
October 5

Topic

Speaker(s)

9:30 - 10:00 am (CT) Attendee check-in
(Please see the ALCF-Workshops Slack #announce channel for connection info)
 
10:00 - 10:10 am (CT)

Welcome [Video]                           

Michael Papka (ALCF Director)
10:10 - 10:30 am (CT)

Intro to SDL Workshop 
(Schedule, speakers, support team, systems, nodes, software, Cobalt)
[SlidesVideo]

Kyle Felker (Argonne)
10:30 am - 12:00 pm (CT) Distributed Deep Learning
 [Video]
Huihuo Zheng, Corey Adams (Argonne)
12:00 - 1:00 pm (CT) Lunch Break   
1:00 - 1:30 pm (CT)

Distributed Deep Learning (cont.)
 

Deploy and Test DeepSpeed on ThetaGPU
[Slides, Video]

Huihuo Zheng, Corey Adams,
Zhen Xie (Argonne)

1:30 - 3:00 pm (CT) Building Data Pipelines [Video] Taylor Childers (Argonne)

Day 2: Wednesday,
October 6

 
9:30 - 10:00 am (CT) Attendee check-in
(Please see the ALCF-Workshops Slack #announce channel for connection info)
 
10:00 am - 12:00 pm (CT) Distributed Hyperparameter Search (HPS) with DeepHyper
[Slides, Video, Video]
Romain Egele, Misha Salim, Sam Foreman (Argonne)
12:00 - 1:00 pm (CT) Lunch Break  
1:00 - 3:00 pm (CT)

Profiling Deep Leaning
(Single-Node TensorFlow, Horovod + TensorFlow)
[Slides, Video]

Murali Emani, Denis Boyda (Argonne)

Day 3: Thursday,
October 7

   
9:30 - 10:00 am (CT) Attendee check-in
(Please see the ALCF-Workshops Slack #announce channel for connection info)
 
10:00 am - 12:00 pm (CT)

Integrating AI and Simulations
[Slides, Video]

Online Learning with SmartSim
[Slides]

Bethany Lusch, Riccardo Balin,
Filippo Simini (Argonne)
12:00 - 1:00 pm (CT) Lunch Break  
1:00 - 2:00 pm (CT)

Neural Architecture Search (NAS) and Uncertainty Quantification (UQ) with DeepHyper
[Slides, Video]

Romain Egele (Argonne)
2:00 - 2:30 pm (CT) Overview of AI Testbeds at Argonne 
[Slides, Video]
Sid Raskar (Argonne)
2:30 - 3:00 pm (CT) Applying for ALCF Allocation Programs
[Slides, Video]
Katherine Riley (Argonne)
3:00 - 3:15 pm (CT) Closing Remarks and Wrap-Up/Next Steps
[Slides]
Ray Loy (Argonne)