Weekly Update: May 10, 2023
In This Message
Meetings & Workshops
– Using Slurm on Frontier (May 18)
– Julia for High-Performance Computing Tutorial (May 24)
– AI Training Series: AI for Science at Scale – Introduction (June 15)
– home.ccs.ornl.gov (May 10)
– Frontier (May 16)
– Darshan loaded by default (May 09)
– Frontier Hackathon – Call for Proposals (Submission deadline is May 26)
– Advanced SYCL Techniques and Best Practices (May 30)
– Introduction to High-Performance Parallel Distributed Computing Using Chapel, UPC++ and Coarray Fortran (July 26-27)
Meeting & Workshops
Using Slurm on Frontier
May 18, 1:00 p.m. EDT
The Oak Ridge Leadership Computing Facility (OLCF) will host a (virtual) “Using Slurm on Frontier” tutorial via Zoom on May 18, 2023 from 1-3 PM EDT. As the name suggests, this session is meant to show new Frontier users how to use Slurm on the Frontier supercomputer. The session will begin with a presentation showing the basics as well as our recommended best practices. After the presentation, existing Frontier users will be encouraged to run their own job steps to make sure their processes and threads are bound to the CPU cores and GPUs they expect. OLCF staff will be available to help during this hands-on portion of the session. This is part of the new Frontier training series.
Julia for High-Performance Computing Tutorial
May 24, 1:00 p.m. EDT
The Oak Ridge Leadership Computing Facility (OLCF) will host a (virtual) “Julia for High-Performance Computing Tutorial” via Zoom on May 24, 2023, from 1-4 PM EDT. OLCF, NERSC, and ALCF users are encouraged to attend. The proposed tutorial introduces participants to the Julia language for high-performance computing (HPC) applications. Julia proposes to fill a gap in the high-performance plus high-productivity space, a dynamic language built on top of LLVM with lightweight interoperability with C and Fortran code and a unified ecosystem for data science and reproducibility.
AI Training Series: AI for Science at Scale – Introduction
June 15, 1:00 – 3:00 pm EDT
Machine learning (ML) is a subset of Artificial Intelligence (AI) that uses statistical learning algorithms to build applications that can automatically learn and improve from its experiences. Deep learning (DL) is a subset of ML that is inspired by the way a human brain filters information (like recognizing patterns and identifying objects). From a science point of view, both ML and DL can be applied to various scientific domains to analyze large datasets, handle noise correction, deal with error classification, and classify features in data.
This training will introduce AI/ML/DL principles used for science in an HPC environment. After learning the “basics”, participants will then be able to apply techniques learned to run hands-on examples using OLCF’s Ascent system.
For more information or to register, see https://www.olcf.ornl.gov/calendar/ai-for-science-at-scale-intro/
- home.ccs.ornl.gov will be unavailable from 4:00 PM until 8:00 PM on Wednesday, May 10
- Frontier will be unavailable from 8:00 AM until 8:00 PM on Tuesday, May 16
Darshan loaded by default
Starting on May 9, 2023, the darshan-runtime module will be loaded by default on Crusher and Frontier. This module will allow users to profile the I/O of their applications with minimal impact. The logs are available to users on the Orion file system in /lustre/orion/darshan/<system>/<yyyy>/<mm>/<dd>. Unloading darshan-runtime is recommended for users profiling their applications with other profilers to prevent conflicts.
Please make a note of this change and contact firstname.lastname@example.org with any issues or questions.
Frontier Hackathon – Call for Proposals
The OLCF, in collaboration with HPE, AMD, and ECP, will be holding a (virtual) Frontier Hackathon on June 28-30, 2023. Application and software teams with existing Frontier allocations are invited to submit proposals to participate in the hackathon. Once the proposal deadline is reached, we will review all proposals and select 10 teams to attend. During the event, the selected teams will work toward their development goals (porting, debugging, optimization, etc.) on Frontier with the help of OLCF, HPE, AMD, and ECP staff. If this particular date does not work for you, fear not – we’ll be holding multiple Frontier Hackathons throughout 2023 so you can submit a proposal for a later date. For more information, or to submit a proposal, please see the additional details below.
Main Frontier Hackathons page: https://www.olcf.ornl.gov/frontier-hackathons/
- Go here for general information on what to expect, who should apply, and a table of hackathons dates.
June 2023 Frontier Hackathon page: https://www.olcf.ornl.gov/frontier-hackathon-june-2023/
- Go here to submit a proposal for the event taking place on June 28-30, 2023.
- The submission deadline is May 26, 2023.
Advanced SYCL Techniques and Best Practices
May 30, 12:00 pm-4:00 pm ET.
Virtual-only event hosted by NERSC
The SYCL programming model means heterogeneous programming using C++ is now more accessible than ever. SYCL uses modern standard C++, and it’s a programming model that lets developers support a wide variety of devices (CPUs, GPUs, FPGAs, and more) from a single code base. Concepts covered in this training include strategies for optimizing code, managing data flow, how to use different memory access patterns, understanding work group sizes, using vectorization, the importance of ND ranges, and making the most of the multiple devices available on your architecture. This training is hosted by NERSC. ALCF and OLCF users are welcome to this training. NERSC training accounts will be provided if needed.
Please use this form to register.
Introduction to High-Performance Parallel Distributed Computing Using Chapel, UPC++ and Coarray Fortran.
Hosted by ECP, NERSC, and OLCF
July 26-27, 12:00pm-3:25pm ET.
In this tutorial, we show how example computations such as heat diffusion, k-mer counting, file processing, and distributed maps can be written to efficiently leverage distributed computing resources in the Chapel, UPC++, and Fortran parallel programming models. This tutorial should be accessible to users with little-to-no parallel programming experience, and everyone is welcome. Current OLCF users with access to Frontier will be able to access a reservation on Frontier to work the examples. Current NERSC users will be able to use Perlmutter. Training accounts on Perlmutter are available for participants who do not have access to either Frontier or are not NERSC users. The examples will also be available in a Docker container and a cloud-based virtual desktop environment for access by any attendee.
Please register here: https://www.exascaleproject.org/event/pgas-2023/