In This Message

Meetings & Workshops
– Nsight Systems and Nsight Compute Profiling Workshop (August 31)
– August 2022 OLCF User Conference call (August 31)
– Data Visualization & Analytics Series: ParaView at OLCF (September 15)
– OLCF GPU Hackathon (Oct 17, 25-27) – deadline to apply is August 25
– OLCF  Annual User Meeting (October 18-19)

Upcoming Downtimes
– Onyx, Jupyter notebooks (August 30, 31)

Meetings & Workshops

Nsight Systems and Nsight Compute Profiling WorkshopAugust 31, 1:00 -3:00 pm EDT

NVIDIA® Nsight™ Systems is a system-wide performance analysis tool designed to visualize an application’s algorithms, help you identify the largest opportunities to optimize, and tune to scale efficiently across any quantity or size of CPUs and GPUs. This OLCF hosted workshop, given by NVIDIA, will show how to use Nsight Systems and Nsight Compute for tracing an HPC application. Users from ALCF, NERSC and OLCF are encouraged to register. If you do not have an account at one of those centers, you are welcome to register watch the lectures but we will not be able to provide you a compute resource for the hands-on parts of the course.

Please find more details and registration here:

August 2022 OLCF User Conference call
August 31, noon – 1:00pm

The August OLCF User Conference Call will be held from noon until 1:00 PM Eastern Time on Wednesday, August 31. During this call, OLCF’s Leah Huk will present an overview of the Andes cluster. Best practices and tasks that are well-suited for Andes (rather than Summit) will be highlighted. For more information, see the event page at

Data Visualization & Analytics Series: ParaView at OLCF
September 15, 1:00 – 3:00 pm EDT

ParaView is an open-source data analysis and visualization application that allows users to quickly build visualizations to analyze their data. Available in both a GUI mode and a batch mode, ParaView is able to analyze extremely large datasets using distributed memory computing resources. This beginner-friendly event will provide an overview of how to access ParaView at OLCF and a tutorial on how to use ParaView to visualize different datasets on Andes.

Only users with moderate security enclave projects (i.e. able to access Summit/Andes) can access the tutorial datasets already downloaded on our filesystem. However, others are welcome to sign up, attend the presentation, and are encouraged to download the public datasets for themselves. Additionally, all examples and datasets will continue to be available after the event for Moderate users to access and work with on their own. For more information or to register, see

OLCF GPU Hackathon 
October 17, 25-27
Do you have an application that needs to be ported to (or optimized on) GPUs? If so, why not work toward your GPU development goals surrounded by experts in the field? This hackathon is a multi-day coding event designed to help teams of three to six developers accelerate their own codes on GPUs using a programming model, or machine learning framework of their choice. Each team is assigned mentors for the duration of the event.

For more information, or to submit a proposal, please visit (The deadline to submit is August 25, 2022)

OLCF Annual User Meeting
October 18-19
The OLCF will hold its annual User Meeting remotely from October 18-19. The purpose of the user meeting is to share selected computational science and engineering achievements emerging from the OLCF’s user programs, to enable interactions among users, to advance the OLCF’s relationships with our user community, and to highlight computational requirements for the future. The 2022 meeting will feature OLCF Facility updates, user-experience talks on Frontier early hardware, deep dive discussions with OLCF staff, and OLCF executive board election results.  For more information or to register, see

Upcoming Downtimes

–   Work will be performed on Onyx from 8:00 AM August 30 until 8:00 AM August 31. During these times, Jupyter notebooks will be restarted, so in-flight processing will be halted but the notebook state will be stored. Otherwise, workflows should migrate between servers in the respective clusters and user impact is not anticipated.