NERSCPowering Scientific Discovery for 50 Years

ALCF Computational Performance Workshop, May 24-26, 2022

May 24, 2022


The 2022 ALCF Computational Performance Workshop will be hosted as an online-only event, and is open to NERSC users. The workshop is designed to help you boost application performance and achieve computational readiness on ALCF systems. The performance tools and topics covered at the workshop are transferrable to NERSC systems too. 

Participants will have an opportunity to:

  • Work with ALCF and industry professionals through collaborative online sessions
  • Participate in guided hands-on exercises
  • Benchmark and debug your application with exclusive reservations on ALCF computing systems
  • Learn how to improve utilization on ThetaGPU and Theta, and prepare for future ALCF computing resources
  • Explore advanced techniques and tools to enhance application performance and expand your data science skills
  • Prepare for a major allocation award (e.g., INCITE, ALCC, ALCF Data Science Program, NERSC ERCAP)

Date and Time: 8 am - 3 pm (Pacific time), Tuesday-Thursday, May 24-26, 2022

The format of this event will be online only. 


The ALCF Computational Performance Workshop is intended for HPC researchers and team members with near-term goals of applying for a major allocation award. Registrants will be reviewed for experience level and goals for attending. It is strongly recommended that registrants provide complete details as to the work they would like to accomplish in coordination with facility staff. There are a limited number of seats available for this virtual event.

For questions or concerns, please email


Day 1, Tuesday, May 24. Preliminary Topics (Subject to change)

  • Onboarding
  • Workshop Overview: Getting Started on Theta and ThetaGPU
  • Overview of TAU
  • Overview of HPC ToolKit
  • Enabling Workflows with Balsam and Parsl
  • Future ALCF Systems
  • Portability with OpenMP
  • Visualizing Your Data
  • Allocation Programs at ALCF
  • Containers

Day 2, Wednesday, May 25

  • NVIDIA Ecosystem
  • Scaling Deep Learning Applications
  • Portability with Kokkos and Raja
  • Data Transfer and Sharing with Globus
  • Data Transfer and Sharing with Eagle
  • Overview of VTune/Advisor

Day 3, Thursday, May 26. Preliminary Topics (Subject to change)

  • Profiling Frameworks with TensorFlow and Pytorch
  • Reduced Precision in TF, Torch
  • File Systems and I/O Performance
  • MPI/Network
  • Portability with SYCL/oneAPI
  • Next Steps - Applying for Your Own Project

Registration and Deadlines

Registration is required for remote participation.  The registration deadlines are:

  • Foreign nationals: April 26, 2022
  • U.S. citizens: May 10, 2022

Please notice that there are a limited number of seats available for this virtual event. Your participation is not confirmed until you receive an acceptance email from ALCF staff.

Register here

Presentation Materials

Slides and videos can be found on the ALCF 2022 Computational Performance Workshop event page