Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “FOCUS PROGRAMS”

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.

At least 1,477 records · Page 82

Advanced Simulation and Computing FY10-FY11 Implementation Plan Volume 2, Rev. 0.5

The Stockpile Stewardship Program (SSP) is a single, highly integrated technical program for maintaining the surety and reliability of the U.S. nuclear stockpile. The SSP uses past nuclear test data along with current and future non-nuclear test data, computational modeling and simulation, and experimental facilities to advance understanding of nuclear weapons. It includes stockpile surveillance, experimental research, development and engineering (D&E) programs, and an appropriately scaled production capability to support stockpile requirements. This integrated national program requires the continued use of current facilities and programs along with new experimental facilities and computational enhancements to support these programs. The Advanced Simulation and Computing Program (ASC) is a cornerstone of the SSP, providing simulation capabilities and computational resources to support the annual stockpile assessment and certification, to study advanced nuclear weapons design and manufacturing processes, to analyze accident scenarios and weapons aging, and to provide the tools to enable stockpile Life Extension Programs (LEPs) and the resolution of Significant Finding Investigations (SFIs). This requires a balanced resource, including technical staff, hardware, simulation software, and computer science solutions. In its first decade, the ASC strategy focused on demonstrating simulation capabilities of unprecedented scale in three spatial dimensions. In its second decade, ASC is focused on increasing its predictive capabilities in a three-dimensional (3D) simulation environment while maintaining support to the SSP. The program continues to improve its unique tools for solving progressively more difficult stockpile problems (focused on sufficient resolution, dimensionality and scientific details); to quantify critical margins and uncertainties (QMU); and to resolve increasingly difficult analyses needed for the SSP. Moreover, ASC has restructured its business model from one that was very successful in delivering an initial capability to one that is integrated and focused on requirements-driven products that address long-standing technical questions related to enhanced predictive capability in the simulation tools. ASC must continue to meet three objectives: (1) Robust Tools - Develop robust models, codes, and computational techniques to support stockpile needs such as refurbishments, SFIs, LEPs, annual assessments, and evolving future requirements; (2) Prediction through Simulation - Deliver validated physics and engineering tools to enable simulations of nuclear weapons performance in a variety of operational environments and physical regimes and to enable risk-informed decisions about the performance, safety, and reliability of the stockpile; and (3) Balanced Operational Infrastructure - Implement a balanced computing platform acquisition strategy and operational infrastructure to meet Directed Stockpile Work (DSW) and SSP needs for capacity and high-end simulation capabilities.

99 GENERAL AND MISCELLANEOUS↗

Simulating Billion-Task Parallel Programs

In simulating large parallel systems, bottom-up approaches exercise detailed hardware models with effects from simplified software models or traces, whereas top-down approaches evaluate the timing and functionality of detailed software models over coarse hardware models. Here, we focus on the top-down approach and significantly advance the scale of the simulated parallel programs. Via the direct execution technique combined with parallel discrete event simulation, we stretch the limits of the top-down approach by simulating message passing interface (MPI) programs with millions of tasks. Using a timing-validated benchmark application, a proof-of-concept scaling level is achieved to over 0.22 billion virtual MPI processes on 216,000 cores of a Cray XT5 supercomputer, representing one of the largest direct execution simulations to date, combined with a multiplexing ratio of 1024 simulated tasks per real task.

Perumalla, Kalyan S↗

GPRA Representation of the Solar Energy Technologies Program

The Solar Program encompasses several technologies in thermal heat and electric markets. The solar buildings component is focused on developing low cost solar hot water and pool heaters to displace fossil-fueled alternatives. For electricity generation, photovoltaics are being improved for both distributed and central generation applications, and the Solar Program is working to accelerate PV adoption through the Million Solar Roofs Initiative. Concentrated Solar Power R&D has also been part of the Solar Program but is not included in the FY04 budget request except at a close-out level. As a result, CSP has not been included in the GPRA04 benefits estimates.

14 SOLAR ENERGY↗

The Material Protection, Control and Accounting Sustainability Program Implementation at the Electrochemical Plant

Joint efforts by the Electrochemical Plant (ECP) in Zelenogorsk, Russia, and the United States Department of Energy National Nuclear Security Administration (US DOE/NNSA) Material Protection, Control and Accounting (MPC&A) Program to upgrade ECP security systems began in 1996. The commissioning of major MPC&A systems at ECP occurred in December 2004. Since that time, the US Project Team (USPT) and ECP personnel have focused jointly on the development and implementation of an enterprise-wide MPC&A Sustainability Program (SP) that address the seven essential MPC&A Program sustainability elements. This paper describes current operational experience at the ECP with the full implementation of the site SP utilizing an earned-value methodology. In support of this site program, ECP has established a Document Control Program (DCP) for sustainability-related documents; developed a robust master Work Breakdown Structure (WBS) that outlines all ECP MPC&A sustainability activities; and chartered an Enterprise-Wide Sustainability Working Group (ESWG) The earned value methodology uses ECP-completed (and USPT-verified) analyses to assess project performance on a quarterly basis. The MPC&A SP, presently operational through a contract between ECP and the Los Alamos National Laboratory (LANL), incorporates the seven essential MPC&A Program sustainability elements and governs all sustainability activities associated with MPC&A systems at ECP. The site SP is designed to ensure over the near term the upgraded MPC&A systems continuous operation at ECP as funding transitions from US-assisted to fully Russian supported and sustained.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Summary Report on the SAE 2016 Range Extenders for Electric Vehicles Symposium

The SAE 2016 Range Extenders for Electric Vehicles Symposium was a 2-day technical meeting focused on the role of advanced internal combustion engines (ICEs) and other novel energy converter technologies for extending the range of electric vehicles (EVs). The first-of-its-kind symposium was notable for focusing solely on the range extender (REx) technologies and not the EVs. The technical program featured presentations from international leaders from industry, government, national laboratories, and academia. The opening keynote presentations covered a broad range of topics including consumer behavior, policy implications, regulatory considerations, and REx architectures as enablers for advanced technologies. The technical sessions focused on an array of REx technologies including conventional ICEs, as well as less conventional or emerging technologies such as microturbines, fuel cells, low-temperature combustion engines, and aluminum-air batteries. The symposium included two panel sessions. The trend toward increasing vehicle electrification and the changing role of ICEs and other auxiliary power unit technologies for use as REx’s is leading to new research and design development needs. The symposium captured the interest of the industry and research communities in exploring the opportunities and challenges associated with REx’s for EVs. This report includes key takeaways, summarized below, and draft notes for each presentation and panel discussion.

33 ADVANCED PROPULSION SYSTEMS↗

Laboratory Directed Research and Development Program Activities for FY 2008.

Brookhaven National Laboratory (BNL) is a multidisciplinary laboratory that maintains a primary mission focus the physical sciences, energy sciences, and life sciences, with additional expertise in environmental sciences, energy technologies, and national security. It is managed by Brookhaven Science Associates, LLC, (BSA) under contract with the U. S. Department of Energy (DOE). BNL's Fiscal year 2008 budget was $531.6 million. There are about 2,800 employees, and another 4,300 guest scientists and students who come each year to use the Laboratory's facilities and work with the staff. The BNL Laboratory Directed Research and Development (LDRD) Program reports its status to the U.S. Department of Energy (DOE) annually in March, as required by DOE Order 413.2B, 'Laboratory Directed Research and Development,' April 19, 2006, and the Roles, Responsibilities, and Guidelines for Laboratory Directed Research and Developlnent at the Department of Energy/National Nuclear Security Administration Laboratories dated June 13, 2006. Accordingly, this is our Annual Report in which we describe the Purpose, Approach, Technical Progress and Results, and Specific Accomplishments of all LDRD projects that received funding during Fiscal Year 2008. BNL expended $12 million during Fiscal Year 2008 in support of 69 projects. The program has two categories, the annual Open Call LDRDs and Strategic LDRDs, which combine to meet the overall objectives of the LDRD Program. Proposals are solicited annually for review and approval concurrent with the next fiscal year, October 1. For the open call for proposals, an LDRD Selection Committee, comprised of the Associate Laboratory Directors (ALDs) for the Scientific Directorates, an equal number of scientists recommended by the Brookhaven Council, plus the Assistant Laboratory Director for Policy and Strategic Planning, review the proposals submitted in response to the solicitation. The Open Can LDRD category emphasizes innovative research concepts with limited management filtering to encourage the creativity of individual researchers. The competition is open to all BNL staff in programmatic, scientific, engineering, and technical support areas. Researchers submit their project proposals to the Assistant Laboratory Director for Policy and Strategic Planning. A portion of the LDRD budget is held for the Strategic LDRD (S-LDRD) category. Projects in this category focus on innovative R&D activities that support the strategic agenda of the Laboratory. The Laboratory Director entertains requests or articulates the need for S-LDRD funds at any time. Strategic LDRD Proposals also undergo rigorous peer review; the approach to review is tailored to the size and scope of the proposal. These Projects are driven by special opportunities, including: (1) Research project(s) in support of Laboratory strategic initiatives as defined and articulated by the Director; (2) Research project(s) in support of a Laboratory strategic hire; (3) Evolution of Program Development activities into research and development activities; and (4) ALD proposal(s) to the Director to support unique research opportunities. The goals and objectives of BNL's LDRD Program can be inferred fronl the Program's stated purposes. These are to (1) encourage and support the development of new ideas and technology, (2) promote the early exploration and exploitation of creative and innovative concepts, and (3) develop new 'fundable' R&D projects and programs. The emphasis is clearly articulated by BNL to be on supporting exploratory research 'which could lead to new programs, projects, and directions' for the Laboratory. We explicitly indicate that research conducted under the LDRD Program should be highly innovative, and an element of high risk as to success is acceptable. To be one of the premier DOE National Laboratories, BNL must continuously foster groundbreaking scientific research. At Brookhaven National Laboratory one such method is through its LDRD Program. This discretionary research and development tool is critical in maintaining the scientific excellence and long-term vitality of the Laboratory. Additionally, it is a means to stimulate the scientific community and foster new science and technology ideas, which becomes a major factor in achieving and maintaining staff excellence and a means to address National needs within the overall mission of the DOE and BNL.

99 GENERAL AND MISCELLANEOUS↗

SRNL Report on Vendor Solicitation and Selective Ion-Electrode Testing

Effective operation of the Hanford Tank Waste Treatment and Immobilization Plant (WTP) will require information collected from sampling and/or monitoring to maintain environmental compliance, product quality, and operating efficiency. The U.S. Department of Energy (DOE) – Office of River Protection (ORP) recognizes the potential long-term benefit of developing real-time, in-line monitoring (RTIM) at the WTP to improve the current strategy of collecting tens of thousands of samples and analyzing them. In-line monitoring can decrease processing time associated with waiting for sample analysis, reduce workers’ exposure to radiation and chemical hazards, decrease secondary waste volumes, potentially improve product quality, and provide production support and enhance overall operational control. The Assistant Manager, WTP chartered an Integrated Technical Team (ITT), which includes personnel from the Savannah River National Laboratory (SRNL), Bechtel National, Inc., and DOE-ORP, to evaluate the capability and potential benefits of replacing routine sample collection requirements with real-time, in-line monitoring. The initial focus of the ITT is the Direct Feed Low Activity Waste (DFLAW) Program. This work scope is being performed in multiple phases. Phase 1 is Requirements and Opportunities Analysis, Phase 2 is Applications and Benefits Determination, and Phase 3 is Qualification and Demonstration. Phase 1 included 6 tasks and was completed by SRNL in 2017. The tasks in Phase 1 were: (1) determining the functional requirements for determining analytes and properties of streams, (2) determining the technical basis for process control, (3) determining the sampling points within the Low Activity Waste (LAW) facility and Effluent Management Facility (EMF), (4) data quality and management, (5) process control challenges, and (6) preparing a Preliminary Analysis Plan. DOE-ORP was briefed on the findings from the Phase 1 tasks and requested that Phase 2 work focus on establishing a basis for a materials balance approach that minimized process sampling. The Phase 2 work scope contains tasks to evaluate a materials balance only approach as well as continuing scoping evaluations for instruments that could be used for RTIM. The Phase 2 work scope was completed by both SRNL and Pacific Northwest National Laboratory (PNNL). The tasks performed in Phase 2 were: 1. Overall program administration and technical oversight 2. Evaluation of a Material Balance Approach, a. Evaluation of uncertainty of this approach compared to uncertainty of current methods b. Evaluation of automating the material balance within the framework of existing control software, c. Development of the methodology to obtain and evaluate data during cold commissioning to support a material balance approach, and d. Development of a framework to utilize during presentation of the revised approach to stakeholders and regulators. 3. Screening Evaluations of Instruments to Supplement the Material Balance Approach, a. AP-105 melter condensate evaluations, and b. Screening testing with simulants. 4. Evaluation of the current DFLAW feed qualification program to identify analytes that could be eliminated, a. Inorganic analytes, b. Radionuclides, and c. Organic analytes. This report describes the results of Task 3b, Screening testing with simulants. The Savannah River National Laboratory requested demonstrations of commercially available analytical instruments by the vendors or laboratory demonstration. The techniques tested included: Raman spectroscopy, Infrared spectroscopy (IR), laser induced breakdown spectroscopy (LIBS), X-Ray Fluorescence Spectroscopy (XRF), and ion-selective electrochemical probes. Additionally, the instrumentation would need to be durable enough to withstand the environment of the plant, preferably with little to no maintenance. The insertion point of the instrument would be dependent upon the instrumentation.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Summary of FY 2008 Geosciences Research

The Department of Energy's research provides a foundation of fundamental knowledge in areas relevant to the Department's missions, and this is particularly true in the area of Geosciences, which has applied programs in multiple other offices. The research support also maintains our stewardship for geosciences research capabilities at the DOE National Laboratories. Excellent fundamental science that can improve and support DOE’s mission needs also provides basic understanding applicable to broader areas of Earth science beyond DOE. The Geosciences Research Program resides within the Division of Chemical Sciences, Geosciences and Biosciences, part of the Office of Basic Energy Sciences of the Office of Science. The participants in this program include researchers at National Laboratories and academic institutions. These activities are formalized by a contract or grant between the Department of Energy and the organization performing the work, providing funds for salaries, equipment, research materials, and overhead. Collaborative work among these institutions is encouraged. The summaries in this document, prepared by the investigators, describe the scope of the individual projects. The Geosciences Research Program includes research in the two broad areas of geophysics and geochemistry. Particular focus areas of interest include rock physics, flow and transport of geologic fluids through porous and fractured media, analytical geochemistry and experimental and theoretical geochemistry. The research is foundational to progress for the Department of Energy's long-range technological needs. Because of the variety of the research needs in the different applied DOE programs, fundamental approaches with multiple potential applications are favored.

58 GEOSCIENCES↗

Performance Assessment of High Efficiency Variable Speed Air-Source Heat Pump in Cold Climate Applications

This project was part of an effort by ComEd's emerging technology program to evaluate the energy saving potential of new energy efficiency technologies. The focus of this technology assessment was to determine energy and peak demand savings potentials of a high efficiency variable speed, air-source, split system heat pump designed for cold climate applications. The results of this technology assessment will be used by CLEAResult to develop a new energy efficiency measure for Commonwealth Edison Company's incentive programs. The project utilized the National Renewable Energy Laboratory's (NREL) Thermal Test Facility to experimentally characterize cooling and heating performance of a high efficiency heat pump split system under varying outdoor climate conditions. The selected climate conditions represented summer and low temperature winter conditions in ComEd's service territory. The laboratory experimentation results were used to develop equipment performance curves required by EnergyPlus hourly simulation engine. Using typical meteorological year 3 weather data for the Chicago O'Hare airport, hourly building simulations (using the EnergyPlus engine) was utilized to estimate the annual energy savings of the high efficiency heat pump in comparison to a standard efficiency unit in the following U.S. Department of Energy (DOE) building codes program energy prototypes: Single-family residence; Strip mall; and Low-rise office.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Coupled Decay Heat and Thermal Hydraulic Capability for Loss-of-Coolant Accident Simulations

As the nuclear energy industry considers ways to achieve improved economics in the current fleet of light-water reactors (LWRs), one possible approach is to operate each cycle for longer durations. This causes a greater portion of the fuel to be burned and reduces the frequency of outages, which ultimately reduces the cost to operate the reactor. However, this also leads to higher burnup fuels than has traditionally been allowed in these reactors. Thus, there are concerns about integrity of high-burnup (HBu) fuel, especially during accident conditions such as loss-of-coolant accidents (LOCAs), as shown by Capps et al.. To investigate these concerns, advanced modeling and simulation capabilities are being leveraged to determine the susceptibility of HBu fuel to fuel fragmentation, relocation, and dispersion (FFRD). Improvements have previously been made to fuel performance capabilities to more accurately model these phenomena; multiphysics simulations have also been conducted to determine the power and burnup histories of the HBu fuel, which are needed as inputs for the fuel performance calculations. Most recently, new statistical approaches have been developed to identify a subset of fuel rods that have greater FFRD susceptibility, reducing the total number of fuel performance simulations required. Prior LOCA simulations have relied on the TRACE systems code, which can model the core and primary loop during accident conditions. TRACE includes many models for various aspects of the primary loop, but two sets of models are important for this report. First, TRACE uses a lumped-fuel approach for modeling the core. This approximates the ~50,000 fuel rods in the core with a much smaller number of rods. The rods can be lumped in various ways as determined by the user. For example, one lumped rod may be used to represent all rods in an assembly, sometimes with an additional rod representing the hottest fuel rod. However, due to runtime constraints and complexity of modeling, a more common approach is to group several assemblies or larger regions of the core into single lumped rods. These lumping schemes apply not only to fuel rods but to flow channels as well. Second, TRACE has several different models for treating decay heat, ranging from pregenerated decay heat curves based on an ANSI/ANS-5.1 standard (hereinafter abbreviated simply as ANSI) to explicit time-dependent heat inputs from the user. None of these models account for differences in isotopics between different rods, which is an approximation the work in this report seeks to eliminate. This report focuses on the implementation of coupled decay heat capabilities in the Virtual Environment for Reactor Applications (VERA) code suite to address a gap identified in previous LOCA simulations. This constitutes an improvement for both the lumped-fuel and decay heat models in TRACE. VERA has been developed to perform high-fidelity, whole-core multiphysics simulations for LWRs. Previously, during the Consortium for Advanced Simulation of LWRs (CASL) program, the emphasis was on providing accurate steady-state analysis—with a secondary focus on reactivity insertion accident (RIA) analysis—to address operational challenges in the nuclear energy industry. Under the Department of Energy (DOE) Nuclear Energy Advanced Modeling and Simulation (NEAMS) program, these capabilities are being extended to a broader range of transient analyses with the goal of quantifying the risk of fuel failures such as FFRD. To properly model such conditions with VERA, decay heat calculations have been integrated with the multiphysics to enable rod-by-rod thermal hydraulic (TH) conditions to be driven by the decay heat in long-running accidents such as LOCAs.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

The Scalable HeterOgeneous Computing (SHOC) Benchmark Suite

Scalable heterogeneous computing systems, which are composed of a mix of compute devices, such as commodity multicore processors, graphics processors, reconfigurable processors, and others, are gaining attention as one approach to continuing performance improvement while managing the new challenge of energy efficiency. As these systems become more common, it is important to be able to compare and contrast architectural designs and programming systems in a fair and open forum. To this end, we have designed the Scalable HeterOgeneous Computing benchmark suite (SHOC). SHOC's initial focus is on systems containing graphics processing units (GPUs) and multi-core processors, and on the new OpenCL programming standard. SHOC is a spectrum of programs that test the performance and stability of these scalable heterogeneous computing systems. At the lowest level, SHOC uses microbenchmarks to assess architectural features of the system. At higher levels, SHOC uses application kernels to determine system-wide performance including many system features such as intranode and internode communication among devices. SHOC includes benchmark implementations in both OpenCL and CUDA in order to provide a comparison of these programming models.

99 GENERAL AND MISCELLANEOUS↗

Collaborating with Utilities to Meet Underserved Community Needs: A Guide to Equitable Commercial Solar and Solar + Storage Deployment

Communities of color and low-income populations dedicate a disproportionate amount of their income to pay for energy. They also experience more frequent and prolonged service disruptions, impacting community health and safety. When designed correctly, commercial solar + storage installations have the potential to ameliorate these issues. However, utility programs created to support such installations are frequently underutilized because communities often lack the time and resources needed to access them. Further, historically, few utilities have had the motivation and incentives or experience to account for this barrier. RMI provided process support to Solar Energy Innovation Network teams, and described their experiences in accessing these clean energy resources to support historically underserved communities in our latest report. The report also outlines the roles of key stakeholders (e.g., utilities, building owners, and community organizations) in supporting program delivery and recommended practices to collaborate to meet community needs. The report focuses on commercial solar and solar + storage resources, including solar + storage microgrids. These technologies can offer benefits to communities who do not own their own properties. When designed to act as a "resilience hub," they can also support communities in times of electricity disruption. The report includes a step-by-step guide to collaborating with utilities to implement programs for historically underserved customers. Utilities can support these projects with financing assistance, modernized rate designs, and, in the deployment phase, technical assistance for system design and interconnection.

14 SOLAR ENERGY↗

XACC: a system-level software infrastructure for heterogeneous quantum–classical computing

Quantum programming techniques and software have advanced significantly over the past five years, with a majority focusing on high-level language frameworks targeting remote REST library APIs. As quantum computing architectures advance and become more widely available, lower-level, system software infrastructures will be needed to enable tighter, co-processor programming and access models. In this work, we present XACC, a system-level software infrastructure for quantum–classical computing that promotes a service-oriented architecture to expose interfaces for core quantum programming, compilation, and execution tasks. Additionally, we detail XACC's interfaces, their interactions, and its implementation as a hardware-agnostic framework for both near-term and future quantum–classical architectures. We provide concrete examples demonstrating the utility of this framework with paradigmatic tasks. Our approach lays the foundation for the development of compilers, associated runtimes, and low-level system tools tightly integrating quantum and classical workflows.

97 MATHEMATICS AND COMPUTING↗

Technical Assistance for Digital Assurance: DistribuTECH Workshop

Idaho National Laboratory, the Department of Energy’s Grid Deployment Office, and other national laboratories are collaborating to ensure the U.S. energy infrastructure is reliable, resilient, and secure. This involves strategically leveraging digital technologies to modernize the grid, enhance its resilience against all hazards and disruptions, and fortify national energy security across a diverse energy portfolio. In this workshop you will learn about Idaho National Lab’s Technical Assistance programs, where organizations will be matched with a national laboratory subject matter expert to focus on their key topical area. The technical assistance offered through this track is designed to be responsive to a rapidly changing regulatory landscape and cutting-edge technologies that enhance grid reliability and efficiency. Users will be guided through a tailored analysis and mitigation program to determine their current security posture and given assistance in evaluating supply chain and protection choices against potential consequences.

Digital assurance↗

A Specification for a Godunov-type Eulerian 2-D Hydrocode, Revision 0

The purpose of this code specification is to describe an algorithm for solving the Euler equations of hydrodynamics in a 2D rectangular region in sufficient detail to allow a software developer to produce an implementation on their target platform using their programming language of choice without requiring detailed knowledge and experience in the field of computational fluid dynamics. It should be possible for a software developer who is proficient in the programming language of choice and is knowledgable of the target hardware to produce an efficient implementation of this specification if they also possess a thorough working knowledge of parallel programming and have some experience in scientific programming using fields and meshes. On modern architectures, it will be important to focus on issues related to the exploitation of the fine grain parallelism and data locality present in this algorithm. This specification aims to make that task easier by presenting the essential details of the algorithm in a systematic and language neutral manner while also avoiding the inclusion of implementation details that would likely be specific to a particular type of programming paradigm or platform architecture.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Performance Portability Evaluation of Fluid-Structure Interaction Simulations on Heterogeneous Platforms

The rapid proliferation of heterogeneous programming languages and multi-vendor hardware has underscored the critical need to evaluate the performance portability of scientific applications. In this work, we present the systematic porting and optimization of a massively parallel fluid-structure interaction code across multiple heterogeneous programming frameworks for deployment on leadership-class supercomputers from major vendors. Our analysis focuses on at-scale performance for simulations involving hundreds of millions of deformable cells, executed on a combination of CPUs and GPUs spanning thousands of nodes on exascale machines. We benchmark the performance of each implementation, highlighting the trade-offs inherent in adopting diverse programming models. Key insights regarding the portability of CUDA on multi-vendor platforms, the superior multi-core CPU performance from SYCL, and architectural considerations on performance optimization are distilled from our experience, offering guidance to other users of high performance computing based on our findings.

Martin, Aristotle [Duke University]↗

A Principled Kernel Testbed for Hardware/Software Co-Design Research

Recently, advances in processor architecture have become the driving force for new programming models in the computing industry, as ever newer multicore processor designs with increasing number of cores are introduced on schedules regimented by marketing demands. As a result, collaborative parallel (rather than simply concurrent) implementations of important applications, programming languages, models, and even algorithms have been forced to adapt to these architectures to exploit the available raw performance. We believe that this optimization regime is flawed. In this paper, we present an alternate approach that, rather than starting with an existing hardware/software solution laced with hidden assumptions, defines the computational problems of interest and invites architects, researchers and programmers to implement novel hardware/software co-designed solutions. Our work builds on the previous ideas of computational dwarfs, motifs, and parallel patterns by selecting a representative set of essential problems for which we provide: An algorithmic description; scalable problem definition; illustrative reference implementations; verification schemes. This testbed will enable comparative research in areas such as parallel programming models, languages, auto-tuning, and hardware/software codesign. For simplicity, we focus initially on the computational problems of interest to the scientific computing community but proclaim the methodology (and perhaps a subset of the problems) as applicable to other communities. We intend to broaden the coverage of this problem space through stronger community involvement.

97 MATHEMATICS AND COMPUTING↗