Program Overview

2020 Exascale Computing Project Annual Meeting


The goal of the 2020 ECP Annual Meeting is to highlight technical accomplishments that are being enabled by interactions and collaborations within the ECP community, including the ECP focus areas, DOE HPC facilities, industry collaborators and participating vendors. In addition to the Keynote and Plenary Sessions, the core of the program will include BoF, Breakout, Panel, Tutorial and Poster Sessions.

Calendar Icon    Program details and final daily agenda are available by clicking here.

=  
Monday Agenda (PDF) Tuesday Agenda (PDF) Wednesday Agenda (PDF) Thursday Agenda (PDF)
All Day Agenda (PDF)




Plenaries

   ● AI for Science
   ● ASCAC Transition of Exascale Computing Project Report
   ● EQSIM: Transforming Earthquake Hazard and Risk Assessment Through Exascale Simulations
   ● Industry Imperative for Exascale


BoF/Breakout/Panel Sessions

The 2020 ECP Annual Meeting offers a diverse set of 45-minute, 60-minute and 90-minute BoF, Breakout and Panel sessions that will take place Tuesday, Wednesday, and Thursday of the Annual Meeting. Participants are encouraged to attend breakouts to learn more about these important topics and areas in the ECP community. Presenters will be discussing accomplishments, requirements sharing, and opportunities for interactions.

See list of BoF/Breakout/Panel sessions below:

   ● Advances in Science and Engineering Enabled by ECP Applications (Part I)
   ● Advances in Science and Engineering Enabled by ECP Applications (Part II)
   ● AMReX and AMReX-Based Applications
   ● App Integration at the Facilities: Lessons Learned so Far
   ● Aurora – Argonne’s Exascale System (NDA Session)
   ● Best Practices for Using Proxy Apps as Benchmarks
   ● CEED: High-Order Methods, Applications, and Performance for Exascale
   ● Checkpointing at Exascale: Software Integration, Use Cases, Challenges
   ● CMakePP: Automating and Simplifying CMake Build Systems
   ● Collaboration Opportunities with the BSSw Fellows
   ● Container Utilization at DOE Compute Facilities
   ● Cultivating Software Sustainability and Productivity through BSSw.io
   ● DAOS (Distributed Asynchronous Object Storage) for Applications
   ● Early Experience of Application Developers with OpenMP Offloading at ALCF, NERSC, and OLCF
   ● ECP 201: Lessons Learned in Preparing Software Technology and Applications for Exascale Systems
   ● ECP and Experimental Science
   ● ECP Center and Application Monitoring - Working Group
   ● ECP Continuous Integration Update
   ● ECP Software Curation through Deployment
   ● ECP SW Deployment Project Team Breakout
   ● Hardware Evaluation for Understanding Network Requirements from Facilities and Software Teams
   ● I/O Performance Addicts
   ● Intelligent Distributed Data Movement for Exascale Computing
   ● Kokkos Ecosystem for Performance Portability: Updates and Plans
   ● Lossy Data Reduction/Compression for ECP Applications
   ● Management of Complex, Heterogeneous, and Hierarchical Memory Resources
   ● MPI BoF
   ● Multiprecision Focus Effort in the ECP Math Library Ecosystem
   ● OLCF ECP Application Engagement
   ● OpenMP Roadmap for Accelerators across DOE Pre-Exascale/Exascale Machines
   ● Overview of the Argonne Aurora Exascale System
   ● Overview of the National Security Applications
   ● PathForward Update – AMD (NDA Session)
   ● PathForward Update – Cray (NDA Session)
   ● PathForward Update – HPE (NDA Session)
   ● PathForward Update – IBM (NDA Session)
   ● PathForward Update – Intel (NDA Session)
   ● PathForward Update – NVIDIA (NDA Session)
   ● Performance Portability at the Intersection of the Exascale Computing Project and the DOE Centers of Excellence
   ● Perlmutter - a Waypoint for ECP Teams
   ● Persistent Memory in the Exascale Memory/Storage Continuum
   ● Preparing Applications for Aurora
   ● Slingshot BoF (NDA Session)
   ● Spack Roundtable
   ● Speed Dating for ECP: 1-on-1 Conversations between AD Teams and Math Library Developers
   ● SUNDIALS Plans and Uses
   ● SYCL Programming Model for Aurora
   ● System Software - Evaluation and Future for Pre-Exascale and Exascale systems
   ● The Compiler Will Help You! The LLVM Compiler Infrastructure in ECP
   ● The Scope and Role of DevOps in HPC
   ● Tools and Techniques for Application Performance Characterization
   ● Understanding Performance with Exa-PAPI

Tutorials

The ECP agenda includes an exciting series of 60-minute, 90-minute, 150-minute, 180-minute, and 330-minute Tutorials offered Tuesday through Thursday of the annual meeting. When planning your schedule for the ECP Annual Meeting, be sure to allow time for you and your teammates to attend some tutorials.

See list of Tutorials below:

   ● A Tutorial Introduction to RAJA Performance Portability Abstractions
   ● Accelerating Your Application I/O with UnifyFS
   ● Achieving High Performance I/O with HDF5
   ● Analyzing ECP Proxy Apps with the Profiling Tool Score-P
   ● Application-Driven Fault-Tolerance for High Performance Distributed Computing
   ● Better Scientific Software Tutorial
   ● CANDLE Hands-on Tutorial
   ● Co-Design Center for Particle Applications (CoPA) Summary
   ● CODAR Hands-on Tutorial
   ● Container Computing for HPC and Scientific Workloads
   ● Customizing Data Services for Fun and Profit
   ● Developing Engaged and Productive Virtual Teams
   ● ECP Continuous Integration Startup Tutorial
   ● ExaGraph: Graph and Combinatorial Methods for Enabling Exascale Applications
   ● ExaLearn Tutorial: Co-Design Center for Exascale Machine Learning Technologies
   ● Exascale I/O with ADIOS: Online Processing, in Situ Visualization and High-Performance Storage I/O
   ● Flux: Using Next-Generation Resource Management and Scheduling Infrastructure for Exascale Workflows
   ● Frontier Tutorial
   ● Getting It Right with Open MPI - Best Practices for Deployment and Tuning of Open MPI
   ● Hands-on with Progress Tracking Cards: A Lightweight Method for Improving Your Software Practices
   ● In Situ Visualization and Analysis with Ascent
   ● Integrating PaRSEC-Enabled Libraries in Scientific Applications
   ● Introduction to the Capabilities and Use of the SUNDIALS Suite of Nonlinear and Differential/Algebraic Equation Solvers
   ● Managing HPC Software Complexity with Spack
   ● Managing Power Efficiency of HPC Applications with Variorum and GEOPM
   ● Memory Movement Orchestration and Topology-Aware Placement with AML
   ● OpenMP 4.5 and 5.0 Tutorial (Offload)
   ● Performance Evaluation Using the TAU Performance System
   ● Performance Tuning with the Roofline Model on GPUs and CPUs
   ● SLATE and MAGMA Linear Algebra & FFT Libraries Tutorial
   ● STRUMPACK / SuperLU: Fast Parallel Direct Linear Solvers and Preconditioners
   ● Umpire: Managing Heterogeneous Memory Resources
   ● UPC++: A PGAS/RPC Library for Asynchronous Exascale Communication in C++
   ● Using HPCToolkit to Measure and Analyze the Performance of GPU-Accelerated Applications