From c7de0944a559986caaade124dc38a6b28111d73f Mon Sep 17 00:00:00 2001 From: Holly Auten Date: Mon, 29 Apr 2024 11:56:00 -0500 Subject: [PATCH] seven new posts --- _posts/2024-03-29-cmvl-new.md | 6 ++++++ _posts/2024-03-29-interpml-new.md | 8 ++++++++ _posts/2024-04-15-ssapy-new.md | 6 ++++++ _posts/2024-04-22-exadis-new.md | 6 ++++++ _posts/2024-04-22-isc24.md | 6 ++++++ _posts/2024-04-24-elcap.md | 6 ++++++ _posts/2024-04-26-dynamicsimccs-new.md | 6 ++++++ 7 files changed, 44 insertions(+) create mode 100644 _posts/2024-03-29-cmvl-new.md create mode 100644 _posts/2024-03-29-interpml-new.md create mode 100644 _posts/2024-04-15-ssapy-new.md create mode 100644 _posts/2024-04-22-exadis-new.md create mode 100644 _posts/2024-04-22-isc24.md create mode 100644 _posts/2024-04-24-elcap.md create mode 100644 _posts/2024-04-26-dynamicsimccs-new.md diff --git a/_posts/2024-03-29-cmvl-new.md b/_posts/2024-03-29-cmvl-new.md new file mode 100644 index 000000000..7d8dc5978 --- /dev/null +++ b/_posts/2024-03-29-cmvl-new.md @@ -0,0 +1,6 @@ +--- +title: "New Repo: CMVL" +categories: new-repo +--- + +The Continuous Monitoring Visualization Library ([CMVL](https://github.com/LLNL/cmvl)) is a broad collection of dashboards, visualizations, queries, and other components for common Logging/SIEM tools such as Elastic or Splunk. It contains useful examples of what can be built allowing for collaboration between DOE laboratories, across the federal government, and with industry partners on the most important components to monitor in an IT environment, particularly including, but not limited to, within HPC systems. diff --git a/_posts/2024-03-29-interpml-new.md b/_posts/2024-03-29-interpml-new.md new file mode 100644 index 000000000..eb370eeb3 --- /dev/null +++ b/_posts/2024-03-29-interpml-new.md @@ -0,0 +1,8 @@ +--- +title: "New Repo: InterpML" +categories: new-repo +--- + +[InterpML](https://github.com/LLNL/interpML) contains Python scripts accompanying the paper "[Leveraging Interpolation Models and Error Bounds for Verifiable Scientific Machine Learning](https://arxiv.org/abs/2404.03586)." Abstract: + +> Effective verification and validation techniques for modern scientific machine learning workflows are challenging to devise. Statistical methods are abundant and easily deployed, but often rely on speculative assumptions about the data and methods involved. Error bounds for classical interpolation techniques can provide mathematically rigorous estimates of accuracy, but often are difficult or impractical to determine computationally. In this work, we present a best-of-both-worlds approach to verifiable scientific machine learning by demonstrating that (1) multiple standard interpolation techniques have informative error bounds that can be computed or estimated efficiently; (2) comparative performance among distinct interpolants can aid in validation goals; (3) deploying interpolation methods on latent spaces generated by deep learning techniques enables some interpretability for black-box models. We present a detailed case study of our approach for predicting lift-drag ratios from airfoil images. Code developed for this work is available in a public GitHub repository. diff --git a/_posts/2024-04-15-ssapy-new.md b/_posts/2024-04-15-ssapy-new.md new file mode 100644 index 000000000..d6a2e89ec --- /dev/null +++ b/_posts/2024-04-15-ssapy-new.md @@ -0,0 +1,6 @@ +--- +title: "New Repo: SSAPy" +categories: new-repo +--- + +The Space Situational Awareness for Python ([SSAPy](https://github.com/LLNL/SSAPy)) is a Python package allowing for fast and precise orbital modeling. All documentation is hosted at [software.llnl.gov/SSAPy/](https://software.llnl.gov/SSAPy/), and the team has established a separate repository for data at [SSAPy-Data](https://github.com/LLNL/SSAPy-Data). diff --git a/_posts/2024-04-22-exadis-new.md b/_posts/2024-04-22-exadis-new.md new file mode 100644 index 000000000..ac087fe90 --- /dev/null +++ b/_posts/2024-04-22-exadis-new.md @@ -0,0 +1,6 @@ +--- +title: "New Repo: ExaDiS" +categories: new-repo +--- + +The Exascale Dislocation Simulator ([ExaDiS](https://github.com/LLNL/exadis)) is a set of software modules written to enable numerical simulations of large groups of moving and interacting dislocations, line defects in crystals responsible for crystal plasticity. By tracking time evolution of sufficiently large dislocation ensembles, ExaDiS predicts plasticity response and plastic strength of crystalline materials. ExaDiS is implemented using the Kokkos framework and built using the CMake build system. diff --git a/_posts/2024-04-22-isc24.md b/_posts/2024-04-22-isc24.md new file mode 100644 index 000000000..1203c9fb0 --- /dev/null +++ b/_posts/2024-04-22-isc24.md @@ -0,0 +1,6 @@ +--- +title: "Join LLNL at ISC24" +categories: event +--- + +LLNL staff will participate in the ISC High Performance Conference ([ISC24](https://www.isc-hpc.com/)) on May 12–16. The event brings together the HPC community—from research centers, commercial companies, academia, national laboratories, government agencies, exhibitors, and more—to share the latest technology of interest to HPC developers and users. [View the calendar of events at LLNL Computing.](https://computing.llnl.gov/about/newsroom/isc24-event-calendar) diff --git a/_posts/2024-04-24-elcap.md b/_posts/2024-04-24-elcap.md new file mode 100644 index 000000000..2eb4bb1e2 --- /dev/null +++ b/_posts/2024-04-24-elcap.md @@ -0,0 +1,6 @@ +--- +title: "LLNL Team Accelerates Multiphysics Simulations with El Capitan Predecessor Systems" +categories: story +--- + +Researchers at LLNL have achieved a milestone in accelerating and adding features to complex multi-physics simulations run on Graphics Processing Units (GPUs), a development that could advance high performance computing and engineering. As LLNL readies for El Capitan, the National Nuclear Security Administration’s first exascale supercomputer, the team’s efforts have centered around the development of MARBL, a next-generation multi-physics code, for GPUs. El Capitan is based on AMD’s cutting-edge MI300A Accelerated Processing Units (APUs), which combines Central Processing Units (CPUs) with GPUs and high-bandwidth memory into a single package, allowing for more efficient resource sharing. In a [recent paper](https://asmedigitalcollection.asme.org/fluidsengineering/article/doi/10.1115/1.4064493/1194096) published by the _Journal of Fluids Engineering_, by harnessing the power of GPUs in El Capitan’s early access machines, the researchers successfully extended MARBL's capabilities to include additional physics crucial for HED physics and fusion modeling. Researchers said the team’s use of performance portability abstraction layers, such as the LLNL-developed [RAJA Portability Suite](https://github.com/LLNL/RAJA) and the [MFEM finite element discretization library](https://mfem.org/) were instrumental in enabling MARBL’s single source code to target multiple GPU/CPU architectures. [Umpire](https://github.com/llnl/umpire), a programming interface that helped alleviate memory constraints on Sierra, also has helped improve codes for El Capitan. RAJA, MFEM, and Umpire are open source. diff --git a/_posts/2024-04-26-dynamicsimccs-new.md b/_posts/2024-04-26-dynamicsimccs-new.md new file mode 100644 index 000000000..530c5a226 --- /dev/null +++ b/_posts/2024-04-26-dynamicsimccs-new.md @@ -0,0 +1,6 @@ +--- +title: "New Repo: DynamicSimCCS" +categories: new-repo +--- + +[DynamicSimCCS](https://github.com/LLNL/DynamicSimCCS) accompanies the manuscript "Advancing carbon capture from bench to pilot scale using dynamic similitude" (link not available yet) accepted for publication at _Cell Reports Physical Science_. The code solves for the one-dimensional concentration and temperature distribution within a carbon capture column using a homogenized absorber model. These solutions are compared to experimental data.