Results for “Exascale applications” — Page 16

DOE OSTI2022

Building a machine learning surrogate model for wildfire activities within a global Earth system model

Abstract. Wildfire is an important ecosystem process, influencing land biogeophysical and biogeochemical dynamics and atmospheric composition. Fire-driven loss of vegetation cover, for example, directly modifies the surface energy budget as a consequence of changing albedo, surface roughness, and partitioning of sensible and latent heat fluxes. Carbon dioxide and methane emitted by fires contribute to a positive atmospheric forcing, whereas emissions of carbonaceous aerosols may contribute to surface cooling. Process-based modeling of wildfires in Earth system land models is challenging due to limited understanding of human, climate, and ecosystem controls on fire counts, fire size, and burned area. Integration of mechanistic wildfire models within Earth system models requires careful parameter calibration, which is computationally expensive and subject to equifinality. To explore alternative approaches, we present a deep neural network (DNN) scheme that surrogates the process-based wildfire model with the Energy Exascale Earth System Model (E3SM) interface. The DNN wildfire model accurately simulates observed burned area with over 90 % higher accuracy with a large reduction in parameterization time compared with the current process-based wildfire model. The surrogate wildfire model successfully captured the observed monthly regional burned area during validation period 2011 to 2015 (coefficient of determination, R2=0.93). Since the DNN wildfire model has the same input and output requirements as the E3SM process-based wildfire model, our results demonstrate the applicability of machine learning for high accuracy and efficient large-scale land model development and predictions.

58 GEOSCIENCES↗

DOE OSTI2025

High-Resolution Modeling of the Gulf of Mexico using E3SM

Coastal ocean modeling is a high priority in the DOE‘s Energy Exascale Earth System Model (E3SM). The goal is to accurately predict the risk of damage to coastal resources and infrastructure due to a changing climate in the coming decades. North American coastal communities are areas of particular interest, as this fits under the topic of US national security and planning assessments in a changing climate. LANL Institutional Computing time for the Tier 1 allocation ”Coastal Ocean and Sea Ice Modeling” have been used for development and testing of numerical methods needed for E3SM coastal applications.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2023

UPC++ v1.0 Programmer’s Guide (Rev. 2023.9.0)

UPC++ is a C++ library that supports Partitioned Global Address Space (PGAS) programming. It is designed for writing efficient, scalable parallel programs on distributed-memory parallel computers. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). The UPC++ control model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. The PGAS memory model additionally provides one-sided RMA communication to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ also features Remote Procedure Call (RPC) communication, making it easy to move computation to operate on data that resides on remote processes. UPC++ was designed to support exascale high-performance computing, and the library interfaces and implementation are focused on maximizing scalability. In UPC++, all communication operations are syntactically explicit, which encourages programmers to consider the costs associated with communication and data movement. Moreover, all communication operations are asynchronous by default, encouraging programmers to seek opportunities for overlapping communication latencies with other useful work. UPC++ provides expressive and composable abstractions designed for efficiently managing aggressive use of asynchrony in programs. Together, these design principles are intended to enable programmers to write applications using UPC++ that perform well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

UPC++ v1.0 Programmer’s Guide (Revision 2022.3.0)

UPC++ is a C++ library that supports Partitioned Global Address Space (PGAS) programming. It is designed for writing efficient, scalable parallel programs on distributed-memory parallel computers. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). The UPC++ control model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. The PGAS memory model additionally provides one-sided RMA communication to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ also features Remote Procedure Call (RPC) communication, making it easy to move computation to operate on data that resides on remote processes. UPC++ was designed to support exascale high-performance computing, and the library interfaces and implementation are focused on maximizing scalability. In UPC++, all communication operations are syntactically explicit, which encourages programmers to consider the costs associated with communication and data movement. Moreover, all communication operations are asynchronous by default, encouraging programmers to seek opportunities for overlapping communication latencies with other useful work. UPC++ provides expressive and composable abstractions designed for efficiently managing aggressive use of asynchrony in programs. Together, these design principles are intended to enable programmers to write applications using UPC++ that perform well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

UPC++ v1.0 Programmer’s Guide, Revision 2023.3.0

UPC++ is a C++ library that supports Partitioned Global Address Space (PGAS) programming. It is designed for writing efficient, scalable parallel programs on distributed-memory parallel computers. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). The UPC++ control model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. The PGAS memory model additionally provides one-sided RMA communication to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ also features Remote Procedure Call (RPC) communication, making it easy to move computation to operate on data that resides on remote processes. UPC++ was designed to support exascale high-performance computing, and the library interfaces and implementation are focused on maximizing scalability. In UPC++, all communication operations are syntactically explicit, which encourages programmers to consider the costs associated with communication and data movement. Moreover, all communication operations are asynchronous by default, encouraging programmers to seek opportunities for overlapping communication latencies with other useful work. UPC++ provides expressive and composable abstractions designed for efficiently managing aggressive use of asynchrony in programs. Together, these design principles are intended to enable programmers to write applications using UPC++ that perform well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

UPC++ v1.0 Programmer’s Guide, Revision 2022.9.0

UPC++ is a C++ library that supports Partitioned Global Address Space (PGAS) programming. It is designed for writing efficient, scalable parallel programs on distributed-memory parallel computers. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). The UPC++ control model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. The PGAS memory model additionally provides one-sided RMA communication to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ also features Remote Procedure Call (RPC) communication, making it easy to move computation to operate on data that resides on remote processes. UPC++ was designed to support exascale high-performance computing, and the library interfaces and implementation are focused on maximizing scalability. In UPC++, all communication operations are syntactically explicit, which encourages programmers to consider the costs associated with communication and data movement. Moreover, all communication operations are asynchronous by default, encouraging programmers to seek opportunities for overlapping communication latencies with other useful work. UPC++ provides expressive and composable abstractions designed for efficiently managing aggressive use of asynchrony in programs. Together, these design principles are intended to enable programmers to write applications using UPC++ that perform well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Assessment of ESM Readiness Level for Exascale HPC

Advancement of Earth System Models (ESMs) is becoming increasingly challenging due to a confluence of factors including increasing model complexity – to more fully represent the earth system, increasing spatial resolution - to achieve higher accuracy by resolving fine-scale dynamical to physical, biological, and chemical processes and their interaction, increasing ensemble size - to more accurately represent predictive uncertainty, and increased computing requirements – to enable more accurate and timely weather predictions and climate projections for societal benefit. The belief by many that computing will take care of itself is no longer valid given the disruptive changes in HPC that are driving up the cost of computing, increasing the difficulty of using emerging HPC effectively, and exposing limits in parallelism, portability and scalability of the ESM applications themselves.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2023

Learning to Predict and Improve Build Successes in Package Ecosystems

Software has become increasingly complex, with a typical application depending on tens or hundreds of packages. Finding compatible versions and build configurations of these packages is challenging. This paper presents a method to learn the likelihood of software build success, and techniques for leveraging this information to guide dependency solvers to better software configurations. We leverage the heavily parameterized package recipes from the Spack package manager to produce a training data set of builds, and we use Graph Neural Networks to learn whether a given package configuration will build successfully or not. We apply our tool to the U.S. Exascale Computing Project’s software stack. We demonstrate its effectiveness in predicting whether a given package will build successfully. We show that our technique can be used to improve the solutions generated by dependency solvers, reducing the need for developers to find working builds by trial and error.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Reconstructing High Resolution ESM Data Through a Novel Fast Super Resolution Convolutional Neural Network (FSRCNN)

In this work, we present the first application of a fast super resolution convolutional neural network (FSRCNN) based approach for downscaling earth system model (ESM) simulations. Unlike other SR approaches, FSRCNN uses the same input feature dimensions as the low resolution input. This allows it to have smaller convolution layers, avoiding over-smoothing, and reducing computational costs. We adapt the FSRCNN to improve reconstruction on ESM data, we term the FSRCNN-ESM. We use high-resolution (~0.25°) monthly averaged model output of five surface variables over North America from the US Department of Energy's Energy Exascale Earth System Model's control simulation. These high-resolution and corresponding coarsened low-resolution (~1°) pairs of images are used to train the FSRCNN-ESM and evaluate its use as a downscaling approach. We find that FSRCNN-ESM outperforms FSRCNN and other super-resolution methods in reconstructing high resolution images producing finer spatial scale features with better accuracy for surface temperature, surface radiative fluxes, and precipitation.

58 GEOSCIENCES↗

DOE OSTI2020

Parallelization and Performance Portability in Hydrodynamics Codes

With the eve of Exascale computing, performance and portability are at the forefront of all scientific codes. Adding more cores and more energy to a system is no longer a sustainable way to achieve performance, and extra effort must now be made to improve performance in all areas of code and code development. Using hydrodynamic codes as a basis, this work explores numerous techniques to achieve performance in different ways. Adaptive mesh refinement (AMR) is a necessary technique to improve memory optimization in mesh-based simulations. However it is invasive and conventionally difficult to integrate into existing applications, so we present a new branch of AMR to create a smooth transition to these optimizations, which not only improves performance, but also greatly reduces developer effort. We introduce the concept of this improvement as Phantom-Cell AMR, and assess theoretically the improvements, as well as present an application of its use. Other work included involves and investigation into an efficient data structure that ensures optimal memory layout for cache performance, with a target of making codes performant and portable across all architectures. All of the work targets both performance and portability, not just on CPU hardware, but specifically across GPU architectures. Parallel performance is key to all of the methods presented, but the research makes a great effort to improve the portability of all applications to prepare for current high performance computing systems and those on the horizon.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2021

xSDK: Building an ecosystem of highly efficient math libraries for exascale

Current efforts to build increasingly powerful computer architectures are opening up new avenues for more complex and higher fidelity simulations coupled with data analytics and learning, leading to new scientific insights and deeper understanding. At one extreme, exascale computers will be much faster than previous computer generations (performing 10 18 operations per second—that is, 1,000 times faster than petascale). To achieve these performance improvements, computer architectures are becoming increasingly complex, with deep memory hierarchies, very high node and core counts, and heterogeneous features such as graphics processing units (GPUs). Such architectural changes impact the full breadth of computing scales, as heterogeneity pervades even current-generation laptops, workstations, and moderate-sized clusters. While emerging advanced architectures provide unprecedented opportunities, they also present significant challenges for developers of scientific applications, such as multiphysics and multiscale codes, who must adapt their software to handle disruptive changes in architectures and new programming models that have not yet stabilized. Developers must consider increasing concurrency while reducing communication and synchronization, and other complexities such as the potential for using mixed precision to leverage the compute power available in low-precision tensor cores. On one hand, developers must implement new scientific capabilities, which in turn increase code complexity. On the other hand, the codes must be ported to new architectures, requiring the inclusion of new programming models and the restructuring of code to achieve good performance. Addressing these issues is beyond the capability of any single person or team—leading to the need for collaboration among many teams, who encapsulate their expertise in reusable software and work together to create sustainable software ecosystems.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

A graphics processing unit accelerated sparse direct solver and preconditioner with block low rank compression

We present the GPU implementation efforts and challenges of the sparse solver package STRUMPACK. The code is made publicly available on github with a permissive BSD license. STRUMPACK implements an approximate multifrontal solver, a sparse LU factorization which makes use of compression methods to accelerate time to solution and reduce memory usage. Multiple compression schemes based on rank-structured and hierarchical matrix approximations are supported, including hierarchically semi-separable, hierarchically off-diagonal butterfly, and block low rank. Here, in this paper, we present the GPU implementation of the block low rank (BLR) compression method within a multifrontal solver. Our GPU implementation relies on highly optimized vendor libraries such as cuBLAS and cuSOLVER for NVIDIA GPUs, rocBLAS and rocSOLVER for AMD GPUs and the Intel oneAPI Math Kernel Library (oneMKL) for Intel GPUs. Additionally, we rely on external open source libraries such as SLATE (Software for Linear Algebra Targeting Exascale), MAGMA (Matrix Algebra on GPU and Multi-core Architectures), and KBLAS (KAUST BLAS). SLATE is used as a GPU-capable ScaLAPACK replacement. From MAGMA we use variable sized batched dense linear algebra operations such as GEMM, TRSM and LU with partial pivoting. KBLAS provides efficient (batched) low rank matrix compression for NVIDIA GPUs using an adaptive randomized sampling scheme. The resulting sparse solver and preconditioner runs on NVIDIA, AMD and Intel GPUs. Interfaces are available from PETSc, Trilinos and MFEM, or the solver can be used directly in user code. We report results for a range of benchmark applications, using the Perlmutter system from NERSC, Frontier from ORNL, and Aurora from ALCF. For a high frequency wave equation on a regular mesh, using 32 Perlmutter compute nodes, the factorization phase of the exact GPU solver is about 6.5× faster compared to the CPU-only solver. The BLR-enabled GPU solver is about 13.8× faster than the CPU exact solver. For a collection of SuiteSparse matrices, the STRUMPACK exact factorization on a single GPU is on average 1.9× faster than NVIDIA’s cuDSS solver.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Climbing the Summit and Pushing the Frontier of Mixed Precision Benchmarks at Extreme Scale

The rise of machine learning (ML) applications and their use of mixed precision to perform interesting science are driving forces behind AI for science on HPC. The convergence of ML and HPC with mixed precision offers the possibility of transformational changes in computational science. The HPL-AI benchmark is designed to measure the performance of mixed precision arithmetic as opposed to the HPL benchmark which measures double precision performance. Pushing the limits of systems at extreme scale is nontrivial -little public literature explores optimization of mixed precision computations at this scale. In this work, we demonstrate how to scale up the HPL-AI benchmark on the pre-exascale Summit and exascale Frontier systems at the Oak Ridge Leadership Computing Facility (OLCF) with a cross-platform design. We present the implementation, performance results, and a guideline of optimization strategies employed for delivering portable performance on both AMD and NVIDIA GPUs at extreme scale.

Lu, Hao↗

DOE OSTI2022

Atmospheric river representation in the Energy Exascale Earth System Model (E3SM) version 1.0

Abstract. The Energy Exascale Earth System Model (E3SM) project is an ongoing, state-of-the-science Earth system modeling, simulation, and prediction project developed by the US Department of Energy (DOE). With an emphasis on supporting the DOE's energy mission, understanding and quantifying how well the model simulates water cycle processes is of particular importance. Here, we evaluate E3SM version 1.0 (v1.0) for its ability to represent atmospheric rivers (ARs), which play significant roles in water vapor transport and precipitation. The characteristics and precipitation associated with global ARs in E3SM at standard resolution (1∘ × 1∘) are compared to the Modern-Era Retrospective analysis for Research and Applications, version 2 (MERRA2). Global patterns of AR frequencies in E3SM show high degrees of correlation (≥0.97) with MERRA2 and low mean absolute errors (MAEs; <1 %) annually, seasonally, and across different ensemble members. However, some large-scale condition biases exist, leading to AR biases – most significant of which are the double intertropical convergence zone (ITCZ), a stronger and/or equatorward-shifted subtropical jet during boreal and austral winters, and enhanced Northern Hemisphere westerlies during summer. By comparing atmosphere-only and fully coupled simulations, we attribute the sources of the biases to the atmospheric component or to a coupling response. Using relationships revealed in Dong et al. (2021), we provide evidence showing the stronger North Pacific jet in winter and the enhanced Northern Hemisphere westerlies during summer, associated with E3SM's double ITCZ and related weaker Atlantic meridional overturning circulation (AMOC), respectively, which are significant sources of the AR biases found in the coupled simulations.

58 GEOSCIENCES↗

DOE OSTI2024

Numerical coupling of aerosol emissions, dry removal, and turbulent mixing in the E3SM Atmosphere Model version 1 (EAMv1) – Part 2: A semi-discrete error analysis framework for assessing coupling schemes

Abstract. Part 1 (Wan et al., 2024) of this study discusses the motivation and empirical evaluation of a revision to the aerosol-related numerical process coupling in the atmosphere component of the Energy Exascale Earth System Model version 1 (EAMv1) to address the previously reported issue of strong sensitivity of the simulated dust aerosol lifetime and dry removal rate to the model's vertical resolution. This paper complements that empirical justification of the revised scheme with a mathematical justification leveraging a semi-discrete analysis framework for assessing the splitting error of process coupling methods. The framework distinguishes the error due to numerical splitting from the error due to the time integration method(s) used for each individual process. Such a distinction results in a framework that provides an intuitive understanding of the causes of the splitting error. The application of this framework to the dust life cycle in EAMv1 confirms (i) that the original EAMv1 scheme artificially strengthens the effect of dry removal processes and (ii) that the revised splitting reduces that artificial strengthening. While the error analysis framework is presented in the context of the dust life cycle in EAMv1, the framework can be broadly leveraged to evaluate process coupling schemes, both in other physical problems and for any number of processes. This framework will be particularly powerful when the various process implementations support a variety of time integration approaches. Whereas traditional local truncation error approaches require separate consideration of each combination of time integration methods, this framework enables evaluation of coupling schemes independent of particular time integration approaches for each process while still allowing for the incorporation of these specific time integration errors if so desired. The framework also explains how the splitting error terms result from (i) the integration of individual processes in isolation from other processes and (ii) the choices of input state and time step size for the isolated integration of processes. Such a perspective has the potential for the rapid development of alternative coupling approaches that utilize knowledge both about the desired accuracy and about the computational costs of individual processes.

58 GEOSCIENCES↗

DOE OSTI2023

Invited Paper: Benchmarking and Optimizing Data Movement on Emerging Heterogeneous Architectures

As supercomputers evolve, nodes are continually increasing in complexity. As a result, each generation of parallel systems brings new performance challenges. For instance, on recent systems inter-node communication has outperformed inter-socket, resulting in poor performance of many node-aware communication optimizations. Communication optimizations are critical for the performance and scalability of parallel applications, but are dependent on the parallel architecture, which varies significantly among recent generations of supercomputers. Furthermore, this paper investigates the performance of various paths of data movement on recent generations of systems, and analyzes the increased complexity of communication, particularly on recent heterogeneous systems. The paper also introduces MPI Advance, a communication library that enables optimizations to be created based on benchmark analysis of each emerging system.

benchmarking↗

DOE OSTI2025

2020 Budget Request for the DOE Computational Science Graduate Fellowship (CSGF) Grant

The Department of Energy Computational Science Graduate Fellowship (DOE CSGF) is essential for addressing the increasingly complex national workforce demands stemming from the growth of computational science and engineering challenges. Computational science and engineering (CSE) takes a multidisciplinary approach that utilizes scientific computing to tackle practical problems and provide technical tools across the spectrum of scientific discovery. The DOE CSGF specifically highlights high-performance computing (HPC) as a critical enabling technology in CSE, driving advancements in science and engineering that are vital to both the DOE and the broader economy. Over the past half-century, HPC has been an essential tool for DOE’s success. During this period, important missions, such as nuclear stockpile stewardship, have turned to HPC as an essential technology. Entire science disciplines have been transformed through the augmentation of scientific observation via HPC. At government laboratories, academic institutions, and in industry, DOE CSGF alumni are helping push traditional HPC boundaries while contributing to discoveries in high-energy physics, quantum information systems, fusion-reactor design, machine learning, additive manufacturing, nano materials for next-generation batteries and transistors, and advanced nuclear reactor modeling. In addition, HPC is used to address national health needs that will eventually point to cures both by helping cancer researchers manage and analyze huge troves of data, by simulating biological mechanisms, and by accelerating drug development. A 2023 report from the ASCAC Subcommittee on American Competitiveness and Innovation to the ASCR office, “Can the United States Maintain Its Leadership in High-Performance Computing?” says of the Program, “The CSGF program provides a barometer for disciplines that will be of interest to future DOE computing. Computational biology, machine learning, and quantum computing are among the subjects that began to swell in the ranks of CSGF applicants before the labs were hiring as high a percentage of employees in these categories.” The explosion of scientific and technological data has heightened the demand for advanced high-performance computing (HPC) to transform these data into meaningful scientific insights. As access to vast amounts of data increases, the fields of Machine Learning and Artificial Intelligence are experiencing a resurgence, enhancing the established practices of computational modeling and simulation. In its September 2020 subcommittee report on "AI/ML, Data Intensive Science, and High-Performance Computing," the DOE Advanced Scientific Computing Advisory Committee (ASCAC) specifically called for a fellowship program to train computational and data scientists to address exascale and data-intensive computing challenges. This integration of empirical and theoretical modeling will increasingly guide federal policymakers in making decisions that impact American society and future generations. It demands a workforce of highly skilled and intellectually agile computational scientists capable of navigating the rapid advancements in scientific computing within the DOE National Laboratory research environment. The DOE CSGF program has consistently addressed this critical need.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Flexible User-Defined Domain Decomposition in Kilometer-Scale E3SM Land Model Simulation

The Energy Exascale Earth System Model (E3SM) Land Model (ELM) has been extended to kilometer-scale (km-ELM) resolutions, enabling high-fidelity simulations of terrestrial processes at 1 km x 1 km grid spacing. In ELM, domain decomposition partitions the computational domain across processors, ensuring efficient parallel execution. Currently, round-robin decomposition is applied, providing a straightforward way to distribute computational workload. As ELM continues evolving at the kilometer-scale (km-scale), particularly with integrating lateral flow modeling, decomposition strategies must also account for the increased workload and data movement. This paper introduces a flexible user-defined domain decomposition framework, allowing users to customize domain partitioning based on application requirements. The impact of different decomposition strategies is evaluated across various applications concerning computation, communication, and I/O. Results demonstrate that while 1D partitioning yields superior I/O performance, k-nearest neighbors (KNN) clustering effectively reduces inter-process communication overhead. This study lays the groundwork for scalable partitioning in large-scale land surface simulations, enhancing next-generation Earth system modeling.

Wang, Dali [ORNL] (ORCID:0000000168065108)↗