top of page

Acerca de

Add-On Workshops

March 7, 2024

There are four add-on workshops available as part of the 17th annual Energy HPC Conference. Each workshop will take place at Rice University's BRC on March 7, 2024 - they will occur simultaneously, so only one workshop can be chosen per registration. All add-on workshops are $75 each and can be purchased alone or in combination with conference tickets. 

​

1.  AI in Energy

2. Best Practices in HPC Systems Management

3. E4S and Programming Toolkits

4. Introduction to Physics-informed Machine Learning with Modulus

 

1. AI in Energy

8:30 am - 2:30 pm | BRC Auditorium

​

Building sustainable futures in High-Performance Computing. Join us to discuss ideas and innovations at the nexus of artificial intelligence and energy.

​

Planning Committee: Denis Akhiyarov, AiKYNETIX; Gibby Dunleavy, Constant Impact; Scott Ferguson, New Era Technology; Keith Gray, Intel; Max Grossman, Cruise; Giewee Hammond, Agellus Tank Robotics; Brianna Hemeyer-Taylor, bp; Tyler Peters, Chevron; Pam Randle, Kinder Morgan; Amy Rueve, Pioneer Natural Resources; Julianna Toms, Halliburton; Xiao-Hui Wu, ExxonMobil

​

8:30 - 9:00 am: Check-in + Breakfast

​

9:00 - 9:30 am: Keynote Address

  • "AI in Energy: Challenges and Opportunities"

    • ​Nadav Cohen, Professor of Computer Science; Co-Founder and Chief Scientist at Imubit 

 

9:30 - 10:30 am: Session 1 - Research and Innovation in AI for Energy

  • Panel | "How Researchers Can Collaborate with AI Panel"

    • Moderator: Prabu Parthasarathy - Global VP - Partner Success, Cognite

    • Panelists:

      • Christy Cardenas, Managing Partner, Grit Ventures; Founder, Grit Labs

      • John Foster, Associate Professor, The Hildebrand Department of Petroleum and Geosystems Engineering, The University of Texas at Austin

      • Vivek Ramavajjala, Founder/CEO, Excarta, Inc.

 

10:30 - 11:00 am: Morning Break + Networking

 

11:00 am - 12:00 pm: Session 2 - Ethics and Responsible AI

  • Panel | "Ensuring Safety and Responsibility in AI-Assisted Coding Panel"

    • Moderator: Jeremy Singer, Product Manager, High Performance Computing, ExxonMobil

    • Panelists:

      • Gibby Dunleavy, Managing Director, Constant Impact

      • Apurva Gala, Principal Science Expert – AI, Shell

      • Jonny Hall, Reservoir Engineering Technology Specialist, bp​​​

 

12:00 - 1:00 pm: Lunch Break + Networking

 

1:00 - 2:30 pm: Session 3 - Advancements in AI Technologies & Application in Energy Industry

  • 20 min Lightning Talks:

    • "Optimization in Energy" - Abi (Abishek) Mukund, Strategy, Imubit

    • "AI for Energy Use Cases with Google Cloud" - Jit Biswas, Principal Architect, Google Cloud

    • "Generative AI for Enhanced Creativity" - Ra Inta, Data Science Manager, Chevron Phillips Chemical

    • "Seismic Data to Subsurface Models with OpenFWI" - Benjamin Consolvo, AI Software Engineering Manager, Intel

 

2:30 pm: Closing Remarks

 

2. Best Practices in HPC Systems Management 

8:00 am - 3:45 pm | BRC Exhibit Hall

Limited to 120 registrants > LIMITED SPOTS AVAILABLE

​

Organizer: Keith Gray, Intel

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

​

8:00 - 8:30 am: Check-in + Breakfast

​

8:30 - 8:35 am: Welcome + Introductions    

  • Keith Gray, Intel

 

8:35 - 9:20 am: TACC Update    

  • John DeSantis, TACC

 

9:20 - 9:40 am: Discovery 5 - From Acceptance to Daily Admin    

  • Raj Gautam, ExxonMobil

 

9:40 - 10:40 am: Panel | Facilities Trends and Best Practices

  • Panelists:

    • David Baldwin, Shell

    • Kent Blancett, bp

    • Donny Cooper, TotalEnergies

    • Shawn Hall, Jump Trading

    • Tim Osborne, ORNL

    • Wade Vinson, NVIDIA (moderator)

 

10:40 - 11:00 am: Break

 

11:00 - 11:30 am: OCP Facilities Projects Update​

  • Michael Schill, Open Compute Project Foundation

 

11:30 am - 12:00 pm: Job Dashboards at ORNL     

  • Tim Osborne, ORNL

 

12:00 - 1:00 pm: Lunch

 

1:00 - 1:30 pm: Standardized Benchmarking and CI/CD Tools for Systems Management

  • Gerard Gorman, Devito Codes

 

1:30 - 2:00 pm: Linux Distribution Planning

  • Jonathon Anderson, CIQ

  • Kent Blancett, bp

​

2:00 - 2:30 pm: Containers and Cluster Management

  • Jonathon Anderson, CIQ

​

2:30 - 3:00 pm: Performance Engineering

  • Ron Cogswell, Shell

  • David Zmick, Jump Trading​

 

3:00 - 3:45 pm: Panel | Challenges, Needs, Plans and Ideas for 2025

  • Panelists:

    • David Baldwin, Shell

    • Kent Blancett, bp

    • Donny Cooper, TotalEnergies

    • John DeSantis, TACC

    • Tim Osborne, ORNL

    • Jeremy Singer, ExxonMobil (moderator)

 

3:45 PM: Close

​

3. E4S and Programming Toolkits

Time 8:30 am - 4:00 pm | BRC 280

Limited to 90 registrants > LIMITED SPOTS AVAILABLE

​

Speakers: 

  • Sameer Shende — Research Professor and the Director of the Performance Research Lab, OACISS, University of Oregon

  • Cristobal A. Barberis — Software Engineer, Adaptive Computing

​

8:30 - 9:00 am: Check-in + Breakfast

9:00 am - 12:00 pm: Session 1

12:00 - 1:00 pm: Lunch

1:00 - 4:00 pm: Session 2

​

The U.S. Department of Energy has developed the Extreme-scale Scientific Software Stack (E4S), a curated Spack based software ecosystem that enables the efficient execution of over 100 HPC and AI/ML applications on diverse platforms that include GPUs from NVIDIA, Intel, and AMD. This tutorial will feature:

​

  • Introduction to the Spack package manager. E4S provides both source builds through the Spack platform and a set of containers that feature a broad collection of HPC software packages. It provides Spack build cache for package binaries, container images, build manifests, and turn-key, from-source builds of popular HPC software packages developed as Software Development Kits (SDKs).

​

  • Using E4S through Adaptive Computing's On Demand Data Center (ODDC) web-based interface to one or more Cloud Service Providers (CSPs) including AWS, Google Cloud, and Microsoft Azure. Hands-on sessions will feature ODDC's performant remote desktop environment based on VNC for launching multi-node jobs on the compute nodes. 

 

  • E4S includes HPC tools including numerical solvers such as PETSc, Trilinos, Sundials, and performance evaluation tools such as TAU, HPCToolkit, and PAPI, and visualization tools such as VisIt and ParaView, and I/O tools such as HDF5, parallel netCDF, and ADIOS2, and applications such as Quantum Espresso, LAMMPS, and OpenFOAM. Hands-on sessions will demonstrate these tools on cloud platforms using ODDC. 

​

  • E4S includes AI/ML tools such as TensorFlow, PyTorch, JAX, LBANN, Scikit-Learn, Pandas, and Keras and other Python tools such as matplotlib, numpy and SciPy that may be launched using integrated Jupyter notebooks. Hands-on sessions will feature interfaces to these tools as well as developing a custom chatbot using Generative AI tools using OpenAI's Python interface. 

​

  • It will describe the community engagements and interactions that led to the many artifacts produced by E4S and how to collaborate with the team and contribute to E4S.

​

​

4. Introduction to Physics-informed Machine Learning with Modulus

8:30 am - 3:00 pm | BRC 1003 (10th Floor Conference Room)

Limited to 30 registrants > LIMITED SPOTS AVAILABLE

​

Speakers: 

  • Pavel Dimitrov, Solutions Architect Manager at NVIDIA        

  • Harpreet Sethi, Solutions Architect at NVIDIA

​

8:30 - 9:00 am: Check-in + Breakfast

9:00 am - 12:00 pm: Session 1

12:00 - 1:00 pm: Lunch

1:00 - 3:00 pm: Session 2

​

High-fidelity simulations in science and engineering are computationally expensive and time-prohibitive for quick iterative use cases, from design analysis to optimization. NVIDIA Modulus, the physics machine learning platform, turbocharges such use cases by building physics-based deep learning models that are 100,000x faster than traditional methods and offer high-fidelity simulation results.

​

Upon completion, you will have an understanding of the various building blocks of Modulus and the basics of physics-informed deep learning. You will also know three different types of Physics ML approaches along with and understanding of their main benefits and drawbacks: PINNs, Neural Operators, Graph Neural Networks.

​

In this hands-on workshop, you'll learn how to:

  • Use the Modulus API

  • Solve data-driven and physics-driven problems using Modulus

  • Utilize techniques that Modulus offers to solve problems ranging from deep learning to modeling multi-physics simulations systems.

 

Prerequisites

 

Tools, Libraries, and Frameworks Used

​

bottom of page