Argonne GPU Hackathon 2021
The Argonne GPU Hackathon is a multi-day 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.
Dates
- April 20, 27-29, 2021
Prerequisites
- Teams are expected to be fluent with the code or project they bring to the event and motivated to make progress during the hackathon.
- No advanced GPU skills required, but teams are expected to know the basics of GPU programming and profiling at the event. A collection of GPU lectures, tutorials, and labs are available for all participants at no fee.
See https://www.gpuhackathons.org/index.php/event/argonne-gpu-hackathon-2021 for eligibility and more information.
Argonne GPU Hackathon 2021
The Argonne GPU Hackathon is a multi-day 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.
Dates
- April 20, 27-29, 2021
Prerequisites
- Teams are expected to be fluent with the code or project they bring to the event and motivated to make progress during the hackathon.
- No advanced GPU skills required, but teams are expected to know the basics of GPU programming and profiling at the event. A collection of GPU lectures, tutorials, and labs are available for all participants at no fee.
See https://www.gpuhackathons.org/index.php/event/argonne-gpu-hackathon-2021 for eligibility and more information.
HDF5 Application Tuning: There is more than one way to skin a cat(fish)
Before returning to application tuning (in part 3), in this second part of the series, we take a closer look at HDF5 performance variability. We highlight the main variability sources, their impact on performance, and considerations for HDF5 container design.
More information about the webinar as well as presentation materials can be found here.
The IDEAS Productivity project, in partnership with the DOE Computing Facilities of the ALCF, OLCF, and NERSC and the DOE Exascale Computing Project (ECP) has resumed the webinar series on Best Practices for HPC Software Developers, which we began in 2016.
As part of this series, we offer one-hour webinars on topics in scientific software development and high-performance computing, approximately once a month. The May webinar is titled Automated Fortran–C++ Bindings for Large-Scale Scientific Applications, and will be presented by Seth Johnson (Oak Ridge National Laboratory). The webinar will take place on Wednesday, May 12, 2021 at 1:00 pm ET.
Abstract:
Although many active scientific codes use modern Fortran, most contemporary scientific software libraries are implemented in C and C++. Providing their numerical, algorithmic, or data management features to Fortran codes requires writing and maintaining substantial amounts of glue code. In the same vein, some projects are actively moving key kernels from Fortran toward C++ to support performance portability models and other rapidly-developing, dynamic programming paradigms. How can a project smoothly connect existing Fortran code to new internal C++ kernels or external C++ libraries? The webinar will introduce SWIG-Fortran, which provides a solution with a wide range of flexibility, including support for performant data transfers, MPI support, and direct translation of C++ features to Fortran interfaces.
The IDEAS Productivity project, in partnership with the DOE Computing Facilities of the ALCF, OLCF, and NERSC and the DOE Exascale Computing Project (ECP) has resumed the webinar series on Best Practices for HPC Software Developers, which we began in 2016.
As part of this series, we offer one-hour webinars on topics in scientific software development and high-performance computing, approximately once a month. The June webinar is titled Using the PSIP Toolkit to Achieve Your Goals – A Case Study at The HDF Group, and will be presented by Elena Pourmal (The HDF Group), Reed Milewicz (Sandia National Laboratories) and Elsa Gonsiorowski (Lawrence Livermore National Laboratory). The webinar will take place on Wednesday, June 9, 2021 at 1:00 pm ET.
Abstract:
Productivity and Sustainability Improvement Planning (PSIP) is a lightweight, iterative workflow that allows software development teams to identify development bottlenecks and track progress toward goals to overcome them. In this talk, we present an overview of the PSIP methodology and toolkit, and describe how the HDF5 Group used PSIP to make improvements in three key areas of their software development process.
Getting Started with E4S for Industry and Agencies Workshop
The Extreme-scale Scientific Software Stack (E4S) project aims to tame both the complexity and portability problems by creating an ecosystem of numerical libraries, runtime systems, and tools that lowers the barrier for entry for the HPC and AI/ML developer communities. E4S is a community effort to provide open source software packages for developing, deploying, and running scientific applications on HPC platforms. It aims to deliver a modular, interoperable, and deployable software stack based on the Spack package manager. The “Getting Started with E4S for Industry and Agencies” workshop was held on June 14-15, 2021. The intended audience was technical people from companies and US government agencies that are considering using E4S in their environment.
Workshop Agenda (Slides)
Day 1: Monday, June 14
8:00 AM – 9:30 AM – Overview E4S, Mike Heroux, Director of ECP Software Technology (Video)
- Who should attend:
- Technical Leaders and developers who may or may not be familiar with E4S;
- Prospective and first-time developers using E4S
- Topics addressed: This overview will discuss the philosophy behind E4S, the E4S components, and why industry should understand and consider adopting some of these components.
10:00 AM – 12:00 PM – Getting Started with E4S for Industry and Agency Tutorial, Sameer Shende (ECP Technical Lead for E4S) and team (Video)
- Who should attend: Prospective and first-time developers of E4S.
- Topics addressed: Hands-on instructions on the E4S build environment. Participants will learn how to incorporate E4S products into codes. Participants may use their own code or one provided.
Day 2: Tuesday, June 15
8:00 AM- 9:00 AM: E4S Q&A (Video)
Opportunity to discuss with Sameer Shende and Mike Heroux any high-level questions on what products to consider for specific industry/agency applications
- Who should attend:
- Technical Leaders and developers who may or may not be familiar with E4S and have questions about applicability to industry problems
- All Tutorial participants
- Topics addressed: High-level questions on what products to consider for specific industry/agency applications, unanswered questions from Day 1
9:00 AM – 11:45 AM: Continuation of hands-on session focusing on AI & ML, GPUs, Sameer Shende, E4S (ECP Technical Lead for E4S) and team (Video)
- Who should attend: Day 1 tutorial participants.
- Topics addressed: Continued hands-on instructions on the E4S environment
Topics addressed in tutorial
- Pantheon demo – David Rogers, LANL
- Performance evaluation tools: TAU
- E4S AI & ML packages
- E4S GPU support
- GPU runtime support: OneAPI, ROCm, and CUDA
- Advanced Spack: E4S Spack build cache
In response to the COVID-19 pandemic and transition to remote work, ECP and the IDEAS Productivity project launched the panel series Strategies for Working Remotely, which explores important topics in this area.
Abstract:
- Many virtual student internship programs across the national labs complex, industry, and academia are underway with students working remotely while geographically dispersed. What does a “virtual internship” lack, and what changes to mentoring or onboarding must be made? How can students get the most out of their internships—what opportunities should they look for, and how can they prepare for challenges? In the ninth installment of the panel discussion series, seasoned scientists offer career advice for students and early career scientists on weathering a pandemic, time management, and how to adjust to long-term changes while working remotely.
Panelists:
- Dorian Arnold, Emory University
- Rebecca Hartman-Baker, Lawrence Berkeley National Laboratory
- Christine Harvey, The MITRE Corporation
- Jay Lofstead, Sandia National Laboratories,
Moderators:
- Ashley Barker, Oak Ridge National Laboratory
- Elaine Raybourn, Sandia National Laboratories
The IDEAS Productivity project, in partnership with the DOE Computing Facilities of the ALCF, OLCF, and NERSC and the DOE Exascale Computing Project (ECP) has resumed the webinar series on Best Practices for HPC Software Developers, which we began in 2016.
As part of this series, we offer one-hour webinars on topics in scientific software development and high-performance computing, approximately once a month. The July webinar is titled Mining Development Data to Understand and Improve Software Engineering Processes in HPC Projects, and will be presented by Boyana Norris (University of Oregon). The webinar will take place on Wednesday, July 7, 2021 at 1:00 pm ET.
Abstract:
The webinar will explore the role of software-related data mining tools in supporting productive development of high-performance scientific software. The webinar will discuss a variety of existing and emerging tools for analyzing code, git, emails, issues, test results, and dependencies, with the long-term goal of improving the understanding of development processes and enhancing developer productivity. The webinar will include specific analysis examples by applying a subset of those tools to ECP projects.
Argonne Training Program on Extreme-Scale Computing
Call for 2021 Applications EXTENDED
The Argonne Training Program on Extreme-Scale Computing (ATPESC) provides intensive, two-week training on the key skills, approaches, and tools to design, implement, and execute computational science and engineering applications on current high-end computing systems and the leadership-class computing systems of the future.
The core of the program will focus on programming methodologies that are effective across a variety of supercomputers and that are expected to be applicable to exascale systems. Additional topics to be covered include computer architectures, mathematical models and numerical algorithms, approaches to building community codes for HPC systems, and methodologies and tools relevant for Big Data applications.
Doctoral students, postdocs, and computational scientists interested in attending ATPESC can review eligibility and application details on the application instructions web page.
The event will be held in the Chicago area. If an in-person meeting is not possible, it will be held as a virtual event.
Note: There are no fees to participate. Domestic airfare, meals, and lodging are provided.
IMPORTANT DATES – ATPESC 2021
- March 5, 2021 (midnight, Anywhere on Earth) – Extended deadline to submit applications
- April 26, 2021 – Notification of acceptance
- May 3, 2021 – Account application deadline
For more information see https://extremecomputingtraining.anl.gov or contact [email protected]
The IDEAS Productivity project, in partnership with the DOE Computing Facilities of the ALCF, OLCF, and NERSC and the DOE Exascale Computing Project (ECP) has resumed the webinar series on Best Practices for HPC Software Developers, which we began in 2016.
As part of this series, we offer one-hour webinars on topics in scientific software development and high-performance computing, approximately once a month. The August webinar is titled Software Engineering Challenges and Best Practices for Multi-Institutional Scientific Software Development, and will be presented by Keith Beattie (Lawrence Berkeley National Laboratory). The webinar will take place on Wednesday, August 4, 2021 at 1:00 pm ET.
Abstract:
Scientific software is increasingly becoming the backbone of obtaining and validating scientific results. This is no longer just the case for traditionally computationally intensive areas but is now true across a wide variety of scientific disciplines. This circumstance elevates how scientific software is developed, independent of the field, to a new level of importance. Further, the multi-institutional nature of many science projects presents unique challenges to how scientific software can be effectively developed and maintained over the long term. In this webinar we present the challenges faced in leading the development of scientific software across a distributed, multi-institutional team of contributors, and we describe a set of best-practices we have found to be effective in producing impactful and trustworthy scientific software.
