2025 Energy HPC Conference

We will present an overview of the Oak Ridge National Laboratory's Artificial Intelligence Initiative, which aims to advance the domains of science, energy, and national security. At the core of this initiative are two fundamental thrusts: transformative science applications and cross-cutting assurance. The application thrust focuses on developing AI methods to accelerate scientific discoveries, while the cross-cutting assurance thrust ensures that AI systems are secure, trustworthy, and energy-efficient. Secure approaches include alignment, privacy preservation, and robustness testing for AI models. Trustworthiness is achieved through validation and verification processes coupled with advanced techniques in uncertainty quantification and causal reasoning. Meanwhile, energy efficiency is prioritized by developing scalable solutions, integrating edge computing technologies, and adopting a holistic co-design approach that optimizes the synergy between software and hardware resources. 

Aurora and Frontier are the nation’s first exascale computing systems for science and engineering, housed at Argonne and Oak Ridge National Laboratories’ Leadership Computing Facility sites, respectively. From the perspective of managing ALCF’s applications-readiness program for Aurora, the Early Science Program, and working within the Applications Integration component of the Department of Energy’s Exascale Computing Program, I will discuss experiences in developing and optimizing applications for these GPU-accelerated architectures. These applications span a wide range of science and engineering domains and include AI and data-intensive computing components—often together in multicomponent workflows. All of the project teams involved were motivated to maintain portability across both exascale platforms, as well as existing pre-exascale platforms. I will discuss their various chosen implementation means to do this, and some cross-cutting best practices.

Speakers: Practitioners and Experts from Industry, Academia, and National Labs

Schedule
8:00 - 8:30 am: Check-in + Breakfast
8:30 - 12:00 pm: Talks
12:00 - 1:00 pm: Lunch
1:00 - 3:45 pm: Talks

Organizers:

• Paul Holzhauer
• Gerard Gorman, Devito Codes
• Fabio Luporini, Devito Codes

Schedule
8:00 - 8:30 am: Check-in + Breakfast
8:30 - 9:00 am: Workshop starts

Organizers:

• Beatrice Riviere
• Matthias Heinkenschloss

Schedule
8:30 - 9:00 am: Check-in + Breakfast
9:00 - 10:00 am: Marta D’Elia (Pasteur Lab, previously at Meta and at Sandia National Lab)
10:00 - 11:00 am: Charbel Farhat (Stanford)
11:00 - 11:20 am: Adrian Celaya (Rice University)
11:20 - 11:40 am: Jonathan Cangelosi (Rice University)
11:40 am - 1:00 pm: Lunch
1:00 - 2:00 pm: Elizabeth Qian (Georgia Tech)
2:00 - 3:00 pm: Benjamin Peherstorfer (NYU)

Course Skill Level: 25% basic content, 25% intermediate content, and 50% advanced content

Materials: Attendees will need to bring their laptop to access materials during the workshop. There will be power, but please charge in advance as some outlets may need to be shared.

Abstract:
The hand-on workshop will present two performance evaluation tools; HPCToolkit and TAU to evaluate and optimize the performance of GPU accelerated HPC and AI applications.

HPCToolkit (https://hpctoolkit.org) is an integrated suite of tools for profiling and tracing of parallel programs on computers ranging from multicore desktop systems to GPU-accelerated supercomputers and cloud platforms. HPCToolkit can measure and analyze executions of fully optimized, dynamically linked parallel applications on tens of thousands of CPU cores and GPUs. It supports multi-lingual codes with external binary-only libraries. It collects sampling based measurements of CPU codes with a controllable overhead. It measures GPU performance using vendor APIs to collect fine-grained measurements using PC sampling or instrumentation and monitors asynchronous GPU operations using activity APIs. HPCToolkit can attribute performance measurements to rich dynamic calling contexts containing procedures, inlined functions, loop nests, and source lines on both CPUs and GPUs.

The TAU Performance System [http://tau.uoregon.edu] is a versatile performance evaluation toolkit supporting both profiling and tracing modes of measurement. It supports performance evaluation of applications running on CPUs and GPUs and supports runtime-preloading of a Dynamic Shared Object (DSO) that allows users to measure the performance without modifying the source code or binary. This tutorial will describe how TAU may be used with MVAPICH and support advanced performance introspection capabilities at the runtime layer. TAU's support for tracking the idle time spent in implicit barriers within collective operations will be demonstrated. TAU also supports event-based sampling at the function, file, and statement level. TAU's support for runtime systems such as CUDA (for NVIDIA GPUs),Level Zero (for Intel oneAPI DPC++/SYCL), ROCm (for AMD GPUs), OpenMP with support for OMPT and Target Offload directives, Kokkos, and MPI allow instrumentation at the runtime system layer while using sampling to evaluate statement-level performance data.

HPCToolkit and TAU will be demonstrated on AWS using the ParaTools Pro for E4S(TM) image. The Extreme-scale Scientific Software Stack (E4S) [https://e4s.io] is a curated, Spack based software distribution of 100+ HPC and AI/ML packages. The Spack package manager is a core component of E4S and it is a platform for product integration and deployment of performance evaluation tools such as HPCToolkit, TAU, DyninstAPI, PAPI, etc. and supports both bare-metal and containerized deployment for CPU and GPU platforms. E4S provides a Spack binary cache and a set of base and full-featured container images with vendor runtimes to support GPU architectures from NVIDIA, Intel, and AMD. E4S is a community effort to provide open-source software packages for developing, deploying, and running scientific applications and tools on HPC platforms.