SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Explore scalability of a resilient task-based domain decomposition preconditioner for elliptic PDEs. Selective reliability to study the impact of different levels of simulated SDC and hard faults. Explore interplay between the application resilience, and the role of the server-client programming model.
We reformulate fundamental numerical problems to run on novel hardware inspired by the brain. Such "neuromorphie hardware consumes less energy per computation, promising a means to augment next-generation exascale computers. However, their programming model is radically different from floating-point machines, with fewer guarantees about precision and communication. The approach is to pass each given problem through a sequence of transformations (algorithmic "reductions") which change it from conventional form into a dynamical system, then ultimately into a spiking neural network. Results for the eigenvalue problem are presented, showing that the dynamical system formulation is feasible.
This report documents the completion of milestone STPRO4-25 Harden and optimize the ROCm based AMD GPU backend, develop a prototype backend for the Intel ECP Path Forward architecture, and improve the existing prototype Remote Memory Space capabilities. The ROCM code was hardened up to the point of passing all Kokkos unit tests - then AMD deprecated the programming model, forcing us to start over in FY20 with HIP. The Intel ECP Path Forward architecture prototype was developed with some initial capabilities on simulators - but plans changed, so that work will not continue. Instead SYCL will be developed as a backend for Aurora. Remote Spaces was improved. Development is ongoing part of a collaboration with NVIDIA.
LANL is working with the Partnership for Advance Combustion Engines (PACE) a DOE sponsored consortium of national laboratories for the research and development for creating smarter engine design tools for sustainable transportation. Much of that effort is focused on heat, mass and momentum transfer within and engine. LANL’s focus in 2019, in particular under PACE has been spray modeling and engine mixture formulation, development of conjugate heat transfer analysis (heat transfer to the environment surrounding the combustion chamber). Our efforts relay on the Fast, Easy, Accurate and Robust Continuum Engineering (FEARCE) software developed at LANL as the next generation of engine modeling software for turbulent reactive and multiphase flow, a 2019 R&D 100 winner. This work is relevant to the DOE Vehicle Technologies Office efforts at addressing national energy security as well. Less dependence on petroleum products leads to greater energy security. By Environmental Protection Agency standards, some vehicles are now reaching 42–50 mpg mark. These are conventional gasoline engines. With continued investment and research into new technical innovations, the potential exists to save more than 4 million barrels of oil per day or approximately $200 to $400 million per day. This would be a significant decrease in emissions and use of petroleum and a very large stimulus to the U.S. economy. Better understanding of fuel injection and fuel–air mixing, thermodynamic combustion losses, and combustion/emission formation processes enhances our ability to minimize fuel use and unwanted emissions. Helping to accomplish this understanding, the FEARCE modeling program is working to provide state-of-theart capability for accurately simulating combustion processes: to have a predictive methodology in a software helping industry and researchers not only meet national goals on fuel usage and emissions, but global goals. In addition, a predictive, robust, and accurate capability for simulating the engine combustion process helps to minimize time and labor for development of new engine technology.
The main goal of this milestone was to help CEED-enabled ECP applications, including ExaSMR, MARBL, ExaWind and ExaAM, to improve their performance and capabilities on GPU systems like Summit and Lassen/Sierra. In addition, the CEED team also worked to: add and improve support for additional hardware and programming models in the CEED software components; release the next version of the CEED software stack, CEED-3.0; and demonstrate performance of libParanumal kernels in libCEED, Nek and MFEM. These additional tasks contributed directly to the main CEED-MS34 goal and will also play an important role in CEED’s future milestones.
Anthony D. Del Genio has served as principal investigator since the inception of ASR in 2009 as co-chair of the Cloud Lifecycle Working Group (CLWG) and in connection with that as a member of the ASR Science and Infrastructure Steering Committee. In this role he has been intimately involved with oversight of existing and proposed ASR Focus Groups (Vertical Velocity, QUICR, Ice Properties) and thematic interest groups (Mesoscale Organized Convection, Warm Low Clouds, Mixed-Phase Clouds, MJO, Entrainment) whose activities fall under the purview of CLWG. The PI actively participates himself in the Mesoscale Organized Convection and MJO groups, is a Co-Investigator on two successfully implemented IOPs (AMIE, MC3E), and was an unfunded Collaborator on the recently completed FASTER project at BNL under the DOE Modeling Program. The PI has also spent the past three years as a member of the ARM Science Board, which reviews proposals to conduct IOPs at the permanent ARM sites and to deploy the ARM Mobile Facilities. Finally, the PI was a participant in the first ARM-European workshop in 2012 that eventually led to the creation of the SGP supersite concept. This report includes an annotated list of published research results.
Scientific computing sometimes involves computation on sensitive data. Depending on the data and the execution environment, the HPC (high-performance computing) user or data provider may require confidentiality and/or integrity guarantees. To study the applicability of hardware-based trusted execution environments (TEEs) to enable secure scientific computing, we deeply analyze the performance impact of AMD SEV and Intel SGX for diverse HPC benchmarks including traditional scientific computing, machine learning, graph analytics, and emerging scientific computing workloads. We observe three main findings: 1) SEV requires careful memory placement on large scale NUMA machines (1x -3.4x slowdown without and 1x -1.15x slowdown with NUMA aware placement), 2) virtualization - a prerequisite for SEV - results in performance degradation for workloads with irregular memory accesses and large working sets (1x -4x slowdown compared to native execution for graph applications) and 3) SGX is inappropriate for HPC given its limited secure memory size and inflexible programming model (1.2x -126x slowdown over unsecure execution). Finally, we discuss forthcoming new TEE designs and their potential impact on scientific computing.
In 2011, Pacific Northwest National Laboratory (PNNL) produced a document known as the Materials Compendium, or Compendium of Material Composition Data for Radiation Transport Modeling, PNNL 15870, Rev. 1, that contains material information useable for modeling purposes for properties of 372 materials. This information is used in several modeling programs used by the radiological/nuclear community, though it is primarily tailored for the Monte-Carlo-N-Particle code produced by Los Alamos National Laboratory. This new document Revision 2 includes a complete review and update of all materials data and references, addressing discrepancies and changes in materials data or references that have occurred since the first revision, an additional 40 materials have been added, primarily newer detector materials developed since the last revision, and isotopic specificity.
High-performance computing relies on ever finer threading. Advances in processor technology include ever greater numbers of cores, hyperthreading, accelerators with integrated blocks of cores, and special vectorized instructions, all of which require more software parallelism to achieve peak performance. Traditional visualization solutions cannot support this extreme level of concurrency. Extreme scale systems require a new programming model and a fundamental change in how we design algorithms. To address these issues we created VTK-m: the visualization toolkit for multi-/many-core architectures.
On behalf of the Advanced Scientific Computing Research (ASCR) program in the US Department of Energy (DOE) Office of Science, we are organizing a Workshop on Reimagining Codesign (ReCoDe). Codesign is the process of jointly designing interoperating components of a computing system—in particular: applications, algorithms, system software, programming models, and the hardware on which they run. The goal is to maximize the overall performance, efficiency, and other desirable qualities of the system as a whole. Codesign is a standard methodology in the embedded-systems community, where space, power, and cost constraints are commonly pitted against execution speed for a tightly constrained feature set. Over the last decade, the DOE has invested in codesign efforts to foster the development of exascale computing systems for broad classes of scientific and engineering applications. The ReCoDe workshop hopes to explore how scientific applications of interest to the DOE can be accelerated through close interactions with hardware designers and software-stack developers, in which all components adapt to each other’s requirements and constraints. We want to answer the question of what are the key tools and methodologies for accomplishing codesign in today’s computing landscape, and what will be the highest impact targets for meeting DOE’s emerging mission requirements. This workshop aims to bring together DOE, industry, and academia to identify opportunities to build on past codesign successes and identify new areas that are either emerging or that may need reimagining for the future. We want to continue to find opportunities that can be pursued as a joint effort and continue to break down the traditional customer/vendor dichotomy with true partnerships. From this work, DOE will benefit from increased application performance relative to what stock hardware or existing general-purpose roadmaps can provide, and vendors will benefit from expanding their hardware’s capabilities to address needs they might have not otherwise anticipated and thereby create more widespread interest in their products. The workshop will be structured around a set of breakout sessions, with every attendee expected to participate actively in the discussions. Afterward, workshop attendees—from DOE, industry, and academia—will produce a report for ASCR that summarizes the findings made during the workshop.
Petroleum refineries increasingly seek to generate fuels with lower carbon intensities (CIs; a measure of life cycle greenhouse gas [GHG] emissions per unit of energy of fuel; well-to-wheel [WTW]) to meet growing demand. Co-processing refers to a process that adds biomass-derived feedstocks to the fossil-based feedstocks of existing petroleum refinery process units. With the use of biofeedstocks, it is expected that co-processed fuels would have lower CIs than their petroleum counterparts without requiring changes in the existing infrastructure for producing, transporting, and using fuels. To quantify the GHG emissions reduction benefits of co-processing, this study uses a linear programming model to simulate petroleum refinery conditions with and without co-processing. The co-processing cases include three biofeedstocks (soy oil, tallow, and used cooking oil or UCO) used as 10 vol.% of the feedstock to a hydrotreater or hydrocracker since these lipid-based feedstocks have favorable properties to be treated in a hydrotreater or hydrocracker. In addition, we considered pyrolysis oil used as 10 vol.% of the feedstock to a fluid catalytic cracking (FCC) unit in the modeled refinery due to its higher oxygen content compared to other lipid-based feedstocks. Life cycle analysis (LCA) using two distinct approaches—process-level energy allocation and a refinery-level marginal approach—has been conducted for each case. The LCA results using process-level allocation show that there are no noticeable changes in emissions or energy use impacts at the facility level. The life cycle GHG emission reductions of co-processing cases are mainly related to the fraction of biogenic carbon embedded in each fuel product. For example, co-processed jet fuels (a mixture of fossil and biogenic fuels) made via hydrotreating or hydrocracking have higher biogenic carbon, which results in jet fuel CI reductions of 3.9%–8.6% compared to the CI of baseline petroleum jet fuels on a WTW basis. However, analysis of co-processed pyrolysis oil in an FCC shows that a higher fraction of biofeedstocks (29%) becomes process emissions (i.e., CO and CO 2 ), mainly due to the oxygenates in pyrolysis oil, and so it generates less renewable fuel than biofeedstocks co-processed via hydrotreating or hydrocracking. Using the refinery-level marginal approach, the changes in energy use and emissions of co-processing cases compared to the petroleum-only baseline case are allocated to the changes in fuel production (assuming renewable fuels). This approach generates the life cycle GHG emission values of co-processed renewable fuels, which are comparable to the CIs of standalone biofuel production pathways. However, as this approach relies on a rough assumption that product yields and emissions from co-processing units on fossil feedstocks remain the same with and without biofeedstock inputs, co-processing cases like FCC pyrolysis oil may generate quite skewed results.
The Exascale Computing Project (ECP) Co-design Center for Particle Applications (CoPA) is addressing the challenges for particle applications to run on upcoming exascale computing architectures. This includes short- and long-range molecular dynamics (MD), particle-in-cell (PIC), and quantum molecular dynamics (QMD). This document addresses $\mathcal{O}(N)$ QMD. The goal of this CoPA team is to develop $\mathcal{O}(N)$ linear scaling methods for QMD optimized for new and next generation architectures. This will provide the capability to simulate larger systems and run for longer timeframes than is possible today. Focus is initially on the density matrix construction. This requires efficient matrix formats and linear algebra operations for distributed architectures. A significant reduction in the wallclock time for an MD timestep is necessary to achieve practical large scale QMD. Equally challenging is to develop scalable distributed implementations. This requires the exploration of efficient data formats (dense, sparse, etc.), data decomposition and distribution across compute nodes using block- and graph-based methods. Traditional programming models such as MPI/OpenMP, hybrid MPI/OpenMP/Accelerator, as well as asynchronous task-based models will be explored to achieve maximum parallelism. Parallelism at the level of a matrix operation, solver, and/or application needs to be explored to gain the most benefit on new and emerging architectures. This report serves as a “living document” of the additions and enhancements throughout the ECP CoPA project for $\mathcal{O}(N)$ QMD. The following sections address the applications we are working with, the quantum chemistry libraries (BML and PROGRESS), and our co-design exploration through the use of a proxy application.
Grid operators can address the inherently stochastic nature of renewables by solving a two-stage stochastic programming model that minimizes the cost of dispatch decisions while accounting for the complex grid dynamics. It is common to use a sample average approximation to estimate the expectation of the second stage costs in this model. However, the large sample count needed for numerical accuracy makes effective modeling large-scale electric grids computationally intractable. We introduce a control variate multi-fidelity estimator for the second-stage recourse that enables high quality dispatch decisions in real-time with a reduced computational burden. We obtain a hierarchy of model fidelities by linearizing the AC power flow system representation to DC power flow, and by relaxing transmission and voltage network constraints. We evaluate the performance of our proposed method on a synthetic grid with 73 buses against a deterministic baseline with persistence forecast and a high-fidelity reference. Our analysis shows a computational speed-up of 7.62x with a minimal loss in accuracy. The multi-fidelity method is well suited to fidelity combinations that use a simplified network topology in their lower fidelity model and is an attractive option for applications where accurate grid modeling needed on a limited computational budget.
Innovation in HPC hardware and adoption of heterogeneous systems has led to a variety of unique programming models. This has led to a challenge for scientific visualization software (and indeed all HPC software) to take full advantage of recent generations of supercomputing. Edge computing, where hardware is specialized for the needs of the particular application, exacerbates the problem. VTK-m has had many successes on this front by providing device-agnostic algorithms that compare favorably to implementations written to specific devices as shown in Table 1. However, VTK-m has focused mostly on GPU and traditional CPU multicore technology. There are numerous processor technologies, both existing and potential future, that are not being addressed by current R&D efforts.
MPI (Message-Passing Interface) is a message-passing library interface specification. All parts of this definition are significant. MPI addresses primarily the message-passing parallel programming model, in which data is moved from the address space of one process to that of another process through cooperative operations on each process. . . MPI is a specification, not an implementation; there are multiple implementations of MPI. This specification is for a library interface; MPI is not a language, and all MPI operations are expressed as functions, subroutines, or methods, according to the appropriate language bindings that, for C and Fortran, are part of the MPI standard. MPI Forum is the organization which is responsible for the MPI Specification.
SYCL is a portable programming model for heterogeneous computing, so it is important to obtain reasonable performance portability of SYCL. Towards the goal of better understanding and improving performance portability of SYCL for machine learning workloads, we have been developing benchmarks for basic operators in deep neural networks (DNNs). These operators could be offloaded to heterogeneous computing devices such as graphics processing units (GPUs) to speed up computation. In this work, we introduce the benchmarks, evaluate the performance of the operators on GPU-based systems, and describe the causes of the performance gap between the SYCL and Compute Unified Device Architecture (CUDA) kernels. We find that the causes are related to the utilization of the texture cache for read-only data, optimization of the memory accesses with strength reduction, shared local memory accesses, and register usage per thread. We hope that the efforts of developing benchmarks for studying performance portability will stimulate discussion and interactions within the community.
The Artificial Intelligence Enhanced Co-Design for Next Generation Microelectronics virtual workshop was held April 4-5, 2023, and attended by subject matter experts from universities, industry, and national laboratories. This was the third in a series of workshops to motivate the research community to identify and address major challenges facing microelectronics research and production. The 2023 workshop focused on a set of topics from materials to computing algorithms, and included discussions on relevant federal legislation and such as the Creating Helpful Incentives to Produce Semiconductors and Science Act (CHIPS Act) which was signed into law in the summer of 2022. Talks at the workshop included edge computing in radiation environments, new materials for neuromorphic computing, advanced packaging for microelectronics, and new AI techniques. We also received project updates from several of the Department of Energy (DOE) microelectronics co-design projects funded in the fall of 2021, and from three of the Energy Frontier Research Centers (EFRCs) that had been funded in the fall of 2022. The workshop also conducted a set of breakout discussions around the five principal research directions (PRDs) from the 2018 Department of Energy workshop report: 1) define innovative material, device, and architecture requirements driven by applications, algorithms, and software; 2) revolutionize memory and data storage; 3) re-imagine information flow unconstrained by interconnects; 4) redefine computing by leveraging unexploited physical phenomena; 5) reinvent the electricity grid through new materials, devices, and architectures. We tasked each breakout group to consider one primary PRD (and other PRDs as relevant topics arose during discussions) and to address questions such as whether the research community has embraced co-design as a methodology and whether new developments at any level of innovation from materials to programming models requires the research community to reevaluate the PRDs developed back in 2018.
On behalf of the Advanced Scientific Computing Research (ASCR) program in the US Department of Energy (DOE) Office of Science, we are organizing a Workshop on Energy-Efficient Computing for Science (EECS). Energy efficiency involves coordination across all the interoperating components of a computing system—in particular, applications, algorithms, system software, programming models, data management, and the hardware on which they run. Looking 10-15 years into the future, the goal is to dramatically lower the energy costs of the computational platforms (from the data center to the edge) serving DOE science while expanding the capabilities of these systems, broadening their applicability to science challenges of interest to DOE and the nation.