Explore the evolving landscape of high-performance computing (HPC) and open-source software initiatives in this informative talk from Lawrence Livermore National Laboratory (LLNL). Learn about the RADIUSS project and the High Performance Computing Innovation Center (HPCIC), which aim to share best practices and foster collaboration in HPC. Discover LLNL's suite of open-source libraries and tools, including Spack, BLT, Umpire, CHAI, and more. Gain insights into the challenges and opportunities in HPC, and understand how LLNL's efforts create a foundation for engagement with potential partners in academia and industry. The speakers discuss the importance of adapting to market forces and physical barriers in the exascale era, highlighting LLNL's leadership in HPC for scientific and national security missions.
RADIUSS and HPCIC: Outreach Through Open-Source Software in the Exascale Era
Inside Livermore Lab via YouTube
Overview
Syllabus
Intro
The HPC computing landscape is changing rapidly
High Performance Computing Innovation Center v2
The HPCIC is partnering with RADIUSS
The RADIUSS project is helping to harden and polish source software for general use
Spack A flexible package manager for HPC
BLT A streamlined CMake build system foundation for HPC software
Umpire An application-focused API for memory management on NUMA and GPU ar
CHAI Optional add-on to RAJA for automating data transfers between mem
LvArray Containers for use in high-performance simulation software
AXOM Application CS Infrastructure
Hypre Highly scalable multilevel solvers and preconditioners
SUNDIALS Adaptive ODE and DAE time integrators and efficient nonlinear sol
libROM Fast & accurate C++ library for reduced order models
XBraid Parallel-in-time multigrid solver software
SAMRAI Structured adaptive mesh refinement applications infrastructure
Caliper A library for always-on performance monitoring
SPOT Performance analysis and history tracking
Flux Next-generation resource management and scheduling framework to addres emerging challenges
Spindle Scalable application start-up
LBANN Livermore Big Artificial Neural Network Toolkit
Ascent Flyweight in-situ visualization and analysis for HPC simulations
ZFP In-memory compression of floating-point and integer arrays
RADIUSS/AWS Tutorial Series
MFEM Tutorial Environment
Plans for future tutorial series & hackathons
Summary and Next Steps
Taught by
Inside Livermore Lab
