SEARCH · Search NASA

Results for “Programming Practices”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 559 records · Page 31

All sources NASA NTRS DOE OSTI

DOE OSTI2023

Harvest and nitrogen effects on bioenergy feedstock quality of grass-legume mixtures on Conservation Reserve Program grasslands

Perennial grass mixtures established on Conservation Reserve Program (CRP) lands can be an important source of feedstock for bioenergy production. This study aimed to evaluate management practices for optimizing the quality of bioenergy feedstock and stand persistence of grass-legume mixtures under diverse environments. A 5-year field study (2008–2012) was conducted to assess the effects of two harvest timings (at anthesis vs after complete senescence) and three nitrogen (N) rates (0, 56, 112 kg N ha -1 ) on biomass chemical compositions (i.e., cell wall components, ash, volatiles, total carbon, and N contents) and the feedstock energy potential, examined by the theoretical ethanol yield (TEY) and the total TEY (i.e., the product of biomass yield and TEY, L ha -1 ), of cool-season mixtures in Georgia and Missouri and a warm-season mixture in Kansas. The canonical correlation analysis (CCA) was used to investigate the effect of vegetative species transitions on feedstock quality. Although environmental variations (mainly precipitation) greatly influenced the management effect on chemical compositions, the delayed harvest after senescence generally improved feedstock quality. In particular, the overall cell wall concentrations and TEY of the warm-season mixtures increased by approximately 7%. Additional N supplies improved the total TEY of both mixtures by ~1.6–4.2 L ha -1 per 1.0 kg N ha -1 input but likely lowered the feedstock quality, particularly for the cool-season mixture. The cell wall concentrations of cool-season mixture reduced by approximately 3%–6%. The CCA results indicated that the increased legume compositions (under low N input) likely enhanced lignin but reduced ash concentrations. This field research demonstrated that with proper management, grass-legume mixtures on CRP lands can provide high-quality feedstock for bioenergy productions.

09 BIOMASS FUELS↗

DOE OSTI2014

Long-Term Monitoring of Utility-Scale Solar Energy Development and Application of Remote Sensing Technologies: Summary Report

In anticipation of increased utility-scale solar energy development over the next 20 to 50 years, federal agencies and other organizations have identified a need to develop comprehensive long-term monitoring programs specific to solar energy development. Increasingly, stakeholders are requesting that federal agencies, such as the U.S. Department of the Interior Bureau of Land Management (BLM), develop rigorous and comprehensive long-term monitoring programs. Argonne National Laboratory (Argonne) is assisting the BLM in developing an effective long-term monitoring plan as required by the BLM Solar Energy Program to study the environmental effects of solar energy development. The monitoring data can be used to protect land resources from harmful development practices while at the same time reducing restrictions on utility-scale solar energy development that are determined to be unnecessary. The development of a long-term monitoring plan that incorporates regional datasets, prioritizes requirements in the context of landscape-scale conditions and trends, and integrates cost-effective data collection methods (such as remote sensing technologies) will translate into lower monitoring costs and increased certainty for solar developers regarding requirements for developing projects on public lands. This outcome will support U.S. Department of Energy (DOE) Sunshot Program goals. For this reason, the DOE provided funding for the work presented in this report.

14 SOLAR ENERGY↗

DOE OSTI2022

Building Analytics Tool Deployment at Scale: Benefits, Costs, and Deployment Practices

Buildings are becoming more data-rich. Building analytics tools, including energy information systems (EIS) and fault detection and diagnostic (FDD) tools, have emerged to enable building operators to translate large amounts of time-series data into actionable findings to achieve energy and non-energy benefits. To expedite data analytics adoption and facilitate technology innovation, building owners, technology developers, and researchers need reliable cost–benefit data and evidence-based guidance on deployment practices. This paper fulfills these needs with the energy use and survey data from a wide-ranging research and industry partnership program that covers thousands of buildings installed with analytics tools. The paper indicates that after two years of implementation, organizations using FDD tools and EIS tools achieved 9% and 3% median annual energy savings, respectively. The median base cost and annual recurring cost for FDD are USD 0.65 per square meter (m2) (USD 0.06 per square foot [ft2]) and USD 0.22 per m2 (USD 0.02 per ft2), and are USD 0.11 per m2 (USD 0.01 per ft2) and USD 0.11 per m2 (USD 0.01 per ft2) for EIS. The common metrics and analyses that are used in the tools to support the discovery of energy efficiency measures are summarized in detail. Two best practice examples identified to maximize the benefits of tool implementation are also presented. Opportunities to advance the state of technology include simplified data integration and management, and more efficient processes for acting on analytics outputs. Compared with previous efforts in the literature, the findings presented in this paper demonstrate the effectiveness of building analytics tools with the largest known dataset.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2023

Evaluating Multimodel Ensemble Seasonal Climate Forecasts on Rangeland Plant Production in the California Annual Grassland

Seasonal precipitation and temperature directly affect total plant production in the California Annual Grassland (CAG). Technological advances have resulted in skillful seasonal climate forecasts (i.e., significant correlations between actual and forecasted climate), which could be input into plant production models to inform stocking and other rangeland management decisions. This study presents a procedure for forecasting plant production in the CAG ecosystem to predict annual plant production for grazing, restoration, or other rangeland management practices using a combination of historical gridMET climate data and seasonal hindcasts (i.e., retrospective forecasts, from the North American Multi-Model Ensemble program). Further, the results of this study first confirmed high forecast skill, throughout the growing season at all sites. We also identified skillful plant production forecasts across most of the growing season at two sites and in three of the seven forecasting months at one study site. Forecasting climate and end-of-year plant production across the growing season at three CAG sites allowed us to identify the places and times in the growing season when forecasting might be most helpful in informing management decisions. Integrating plant production forecasting into rangeland management practices could significantly improve rangeland management outcomes. These procedures provide a user guide for creating plant production forecasts for any given area of interest and may be applicable across a wide range of other agricultural and rangeland management systems.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2015

Accelerating k-NN Algorithm with Hybrid MPI and OpenSHMEM

Machine Learning algorithms are benefiting from the continuous improvement of programming models, including MPI, MapReduce and PGAS. k-Nearest Neighbors (k-NN) algorithm is a widely used machine learning algorithm, applied to supervised learning tasks such as classification. Several parallel implementations of k-NN have been proposed in the literature and practice. However, on high-performance computing systems with high-speed interconnects, it is important to further accelerate existing designs of the k-NN algorithm through taking advantage of scalable programming models. To improve the performance of k-NN on large-scale environment with InfiniBand network, this paper proposes several alternative hybrid MPI+OpenSHMEM designs and performs a systemic evaluation and analysis on typical workloads. The hybrid designs leverage the one-sided memory access to better overlap communication with computation than the existing pure MPI design, and propose better schemes for efficient buffer management. The implementation based on k-NN program from MaTEx with MVAPICH2-X (Unified MPI+PGAS Communication Runtime over InfiniBand) shows up to 9.0% time reduction for training KDD Cup 2010 workload over 512 cores, and 27.6% time reduction for small workload with balanced communication and computation. Experiments of running with varied number of cores show that our design can maintain good scalability.

Lin, Jian↗

DOE OSTI2022

Best Practices for NERSC Training

The National Energy Research Supercomputing Center (NERSC) at Lawrence Berkeley National Laboratory (LBNL) organizes approximately 20 training events per year for its 8,000 users from 800 projects, who have varying levels of High Performance Computing (HPC) knowledge and familiarity with NERSC's HPC resources. Due to the novel circumstances of the pandemic, NERSC began transforming our traditional smaller-scale, on-site training events to larger-scale, fully virtual sessions in March 2020. We treated this as an opportunity to try new approaches and improve our training best practices. This paper describes the key practices we have developed since the start of this transformation, including considerations for organizing events; collaboration with other HPC centers and the DOE ECP Program to increase reach and impact of events; targeted emails to users to increase attendance; efficient management of user accounts for computational resource access; strategies for preventing Zoombombing; streamlining the publication of professional-quality, closed-captioned videos on the NERSC YouTube channel for accessibility; effective communication channels for Q&A; tailoring training contents to NERSC user needs via close collaboration with vendors and presenters; standardized training procedures and publishing of training materials; and considerations for planning HPC training topics. Additionally, most of these practices will be continued after the pandemic as effective norms for training.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2018

Changing Institutional Procurement Behavior to Achieve Energy Savings

Although federal policies regarding energy-efficient product procurement (EEPP) are long-standing and well-established, federal buyers do not typically request energy-efficient products when making purchases. This is a large missed opportunity: full compliance could save the U.S. federal government roughly $500 million in energy cost annually. In 2016, the U.S. Department of Energy’s Federal Energy Management Program (FEMP) began a new set of program activities to increase federal compliance with these procurement requirements. Program interventions included increased communication with contracting officers during the solicitation process; development of enhanced, targeted training; and collection and dissemination of procurement best practices across the federal sector. We have collected and analyzed ~2,500 solicitations from over 40 different federal agencies to evaluate the influence of these interventions. Compared to pre-intervention years (FY15 and FY11), annual compliance rates for FY16 and FY17 increased by about 10%. This paper provides an overview of the data collection process, analysis framework, intervention method, and results of our analysis. We discuss strategies to increase compliance rates by examining institutional factors that drive procurement behavior. By combining data collection with adjustment to program implementation, we have created an iterative process that is having a demonstrable effect in improving the impact of a long-running program.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2025

Proposed Classifications of Remote Operations for Nuclear Reactors Based on Physical and Cybersecurity Considerations

The incorporation of remote operations into reactor operations is a topic of high interest among advanced and small modular reactor (A/SMR) vendors, with some considering it essential to the success of their business models. However, remote operations are a concept novel to the nuclear industry. While various technical aspects of remote operations have been explored, a significant gap remains in understanding the security implications of integrating remote operations into reactor designs, particularly concerning the security requirements for remote-operations facilities and infrastructure. This report aims to address this gap by first defining classes of remote operation based on the extent of remote access to reactor control systems and grounded in the existing regulatory framework with compatible terminology. Secondly, the report outlines the physical and cybersecurity requirements applicable to remote-operations facilities and infrastructure at each defined class. These requirements are based on existing licensing frameworks provided by 10 Code of Federal Regulations (CFR) Part 50 and 10 CFR Part 52, as well as the upcoming A/SMR licensing framework in the proposed Part 53. The assessment focuses specifically on security regulations, such as 10 CFR Part 73, which includes provisions for both cybersecurity (§ 73.54) and physical security (§ 73.55). This report proposes five classes of remote reactor operations. Class 1 involves remote monitoring only, with no control over reactor systems. Class 2 allows for the remote issuance of allowlisted commands to the reactor facility. Class 3 extends control to non-safety-significant, non-safety-related, or not important to safety systems and equipment. Class 4 permits remote control of safety-significant systems. Finally, Class 5 allows remote control of safety-related systems. It is important to note that these classes were defined purely with functionality in mind, without considering the practicality or feasibility of implementation for each class under current or upcoming regulatory guidance. The intention behind this approach is to enable an assessment of which security requirements apply to each class, allowing readers to evaluate the implementation possibilities for their specific use cases. Following the definition of remote-operation classes, the report assesses the specific physical and cybersecurity requirements applicable to the remote-operations facility and infrastructure within each defined class. This includes defining the types and locations of operators that are possible at each class of operation and, based on operator type and location, as well as functionality within each class, outlining the physical and cybersecurity requirements. By detailing the security requirements by class, the report provides readers with the information needed to determine the type of security program they may need to implement for their desired concept of operation. The next contribution of this report was to assess the practicality of implementing each proposed class of remote operations based upon the security requirement assessment. In short, three of the five proposed remote-operation classes were found to possibly have a practical path forward to implementation under the U.S. regulatory framework. Class 1 remote operations are currently in use in the U.S. while Class 2 and 3 remote operations may be logistically possible to implement under the U.S. regulatory framework. The final two Classes, 4 and 5, would likely be logistically difficult, if not infeasible to implement within the current U.S. physical- and cybersecurity regulatory framework. Given the results of the feasibility assessment, an example architecture is proposed for both Class 2, remote allowlisted commands, and Class 3, remote control of non-safety systems as well as security implication assessments of each architecture. These example implementations are not meant to be prescriptive in terms of how Class 2 or Class 3 remote operations should be deployed; instead, they are intended to be informative to stakeholders on how Class 2 or Class 3 could potentially be applied in order to inform their system design. An example architecture for Class 1 remote monitoring was not provided as Class 1 in already in use in U.S. nuclear operations. Example architectures for Class 4 and Class 5 were not provided due to their assessment of being likely infeasible to implement. The final contribution is an assessment of the physical- and cybersecurity implications of introducing autonomous operations into an A/SMR. What was found was that the security implications can be separated into two cases. Autonomous operations supported by SSCs located only at the reactor site, and autonomous operations supported by SSCs outside of the reactor site. For the first case, the introduction of autonomous systems will likely not change the facility’s requirement to comply with existing cyber and physical security regulation

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2015

Practical Implementation of Failure Mode and Effects Analysis for Safety and Efficiency in Stereotactic Radiosurgery

Purpose: To improve the safety and efficiency of a new stereotactic radiosurgery program with the application of failure mode and effects analysis (FMEA) performed by a multidisciplinary team of health care professionals. Methods and Materials: Representatives included physicists, therapists, dosimetrists, oncologists, and administrators. A detailed process tree was created from an initial high-level process tree to facilitate the identification of possible failure modes. Group members were asked to determine failure modes that they considered to be the highest risk before scoring failure modes. Risk priority numbers (RPNs) were determined by each group member individually and then averaged. Results: A total of 99 failure modes were identified. The 5 failure modes with an RPN above 150 were further analyzed to attempt to reduce these RPNs. Only 1 of the initial items that the group presumed to be high-risk (magnetic resonance imaging laterality reversed) was ranked in these top 5 items. New process controls were put in place to reduce the severity, occurrence, and detectability scores for all of the top 5 failure modes. Conclusions: FMEA is a valuable team activity that can assist in the creation or restructuring of a quality assurance program with the aim of improved safety, quality, and efficiency. Performing the FMEA helped group members to see how they fit into the bigger picture of the program, and it served to reduce biases and preconceived notions about which elements of the program were the riskiest.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

DOE OSTI2022

Low refractive index OLEDs for practical high-efficiency outcoupling. Final report

Improving OLED light extraction is a significant on-going challenge for the OLED lighting industry. The goal of this program was to explore the potential to improve internal light extraction in a manufacturable way by lowering the refractive index of the OLED stack through dilution with an electrically-inert, low refractive index molecule. We accomplished this goal, identifying a small molecule that can be co-evaporated in the hole transport layer of OLEDs at high loading to reduce the refractive index by 0.2-0.3 refractive index units without degrading charge transport. Implementing this dilution molecule in the hole transport layer of a range of single and multi-stack OLEDs, we demonstrated the ability to improve their internal light extraction by up to ~35% without increasing their drive voltage or reducing their operational lifetime. Importantly, these findings extend to state-of-the-art device architectures with proprietary OLED materials.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2014

Beyond the Active Site: The Impact of the Outer Coordination Sphere on Electrocatalysts for Hydrogen Production and Oxidation

Hydrogenase enzymes provide inspiration for investigations of molecular catalysts utilizing structural and functional mimics of the active site. However, the resulting active site mimics cannot match the combination of high rates and low overpotentials of the enzyme, suggesting that the rest of the protein scaffold, i.e., the outer coordination sphere, is necessary for the efficiency of hydrogenase. Therefore, inclusion of outer coordination sphere elements onto molecular catalysts may enable us to achieve and ultimately surpass the overall enzymatic efficiency. In an effort to identify and include the missing enzymatic features, there has been recent effort to understand the effect of outer coordination sphere elements on molecular catalysts for hydrogen oxidation and production. Our focus has been to utilize amino acid or peptide based scaffolds on an active functional mimic for hydrogen oxidation and production, [Ni(PR2NR’2)2]2+. This bottom-up approach, i.e, building an outer coordination sphere around a functional molecular catalyst, has allowed us to evaluate individual contributions to catalysis, including enhancing proton movement, concentrating substrate and introducing structural features to control reactivity. Collectively, these studies have resulted in catalysts that can operate faster, can operate at lower overpotentials, have enhanced water solubility, and/or can provide more stability to oxygen or extreme conditions such as strongly acidic or basic conditions than their unmodified parent complexes. Common mechanisms have yet to be defined to predictably control these processes but our growing knowledge in this area is essential for the eventual mimicry of enzymes for developing efficient molecular catalysts for practical use. This account reviews previously published work supported by the US DOE Basic Energy Sciences (BES), Physical Bioscience program, the Office of Science Early Career Research Program through the USDOE, BES, the Center for Molecular Electrocatalysis, an Energy Frontier Research Center funded by the US DOE, Office of Science, Office of BES. Part of the research was conducted at the W.R. Wiley Environmental Molecular Sciences Laboratory, a national scientific user facility sponsored by U.S. DOE’s Office of Biological and Environmental Research program located at Pacific Northwest National Laboratory (PNNL). PNNL is operated by Battelle for the U.S. Department of Energy.

bioinspired↗

DOE OSTI2016

SU-F-J-211: Scatter Correction for Clinical Cone-Beam CT System Using An Optimized Stationary Beam Blocker with a Single Scan

Purpose: X-ray scatter photons result in significant image quality degradation of cone-beam CT (CBCT). Measurement based algorithms using beam blocker directly acquire the scatter samples and achieve significant improvement on the quality of CBCT image. Within existing algorithms, single-scan and stationary beam blocker proposed previously is promising due to its simplicity and practicability. Although demonstrated effectively on tabletop system, the blocker fails to estimate the scatter distribution on clinical CBCT system mainly due to the gantry wobble. In addition, the uniform distributed blocker strips in our previous design results in primary data loss in the CBCT system and leads to the image artifacts due to data insufficiency. Methods: We investigate the motion behavior of the beam blocker in each projection and design an optimized non-uniform blocker strip distribution which accounts for the data insufficiency issue. An accurate scatter estimation is then achieved from the wobble modeling. Blocker wobble curve is estimated using threshold-based segmentation algorithms in each projection. In the blocker design optimization, the quality of final image is quantified using the number of the primary data loss voxels and the mesh adaptive direct search algorithm is applied to minimize the objective function. Scatter-corrected CT images are obtained using the optimized blocker. Results: The proposed method is evaluated using Catphan@504 phantom and a head patient. On the Catphan©504, our approach reduces the average CT number error from 115 Hounsfield unit (HU) to 11 HU in the selected regions of interest, and improves the image contrast by a factor of 1.45 in the high-contrast regions. On the head patient, the CT number error is reduced from 97 HU to 6 HU in the soft tissue region and image spatial non-uniformity is decreased from 27% to 5% after correction. Conclusion: The proposed optimized blocker design is practical and attractive for CBCT guided radiation therapy. This work is supported by grants from Guangdong Innovative Research Team Program of China (Grant No. 2011S013), National 863 Programs of China (Grant Nos. 2012AA02A604 and 2015AA043203), the National High-tech R&D Program for Young Scientists by the Ministry of Science and Technology of China (Grant No. 2015AA020917)

60 APPLIED LIFE SCIENCES↗

DOE OSTI2017

XPRESS: eXascale PRogramming Environment and System Software

The XPRESS Project is one of four major projects of the DOE Office of Science Advanced Scientific Computing Research X-stack Program initiated in September, 2012. The purpose of XPRESS is to devise an innovative system software stack to enable practical and useful exascale computing around the end of the decade with near-term contributions to efficient and scalable operation of trans-Petaflops performance systems in the next two to three years; both for DOE mission-critical applications. To this end, XPRESS directly addresses critical challenges in computing of efficiency, scalability, and programmability through introspective methods of dynamic adaptive resource management and task scheduling.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2018

Cost-effective Manufacturing and Morphological Stabilization of Nanostructured Cathodes for Commercial SOFCs

The overall goals of the program were to develop methods for preparing and stabilizing SOFC electrodes by infiltration that can be manufactured on a practical scale, then transfer those methods to cells in stacks. Cathodes prepared by infiltration of nanoparticles into a porous scaffold of the electrolyte have demonstrated outstanding electrochemical performance and mechanical strength but are difficult to manufacture due to the large number of infiltration steps required to produce a conductive composite and show poor stability due to sintering of the nanoparticles. At Penn, the main goal was to develop a fabrication method that uses only a small number of infiltration steps. Infiltration with molten salts and electrodeposition were pursued but deemed impractical. A method in which an electronically conductive scaffold consisted of a composite of a mixed conductor, LSF (La 0.9 Sr 0.1 FeO 3 ), with yttria-stabilized zirconia (YSZ) was found to be promising. While some cation mixing between LSF and YSZ was observed during calcination, there was no evidence for new phases in XRD and incorporation of Zr into the perovskite could be minimized by reducing the Sr content of the LSF. To stabilize nanoparticles infiltrated into the LSF-YSZ composite scaffold, the USC team examined the effects of Atomic Layer Deposition with ZrO 2 and infiltrated nanoparticles. Significantly enhanced stability was observed with the infiltrated nanoparticles. The FuelCell Energy team prepared 4”x4” cells in which an LSF-YSZ composite scaffold was cofired with the YSZ electrolyte and infiltrated with LSCF nanoparticles. Apparently because the LSF-YSZ scaffold layer was very thin in this case, the cation mixing that occurred during calcination resulted in a poor scaffold-electrolyte interface which caused reduced performance.

36 MATERIALS SCIENCE↗

DOE OSTI2022

Cost-effective Manufacturing and Morphological Stabilization of Nanostructured Cathodes for Commercial Solid Oxide Fuel Cells (SOFCs) [Slides]

30 DIRECT ENERGY CONVERSION↗

DOE OSTI2021

Microstructural-Level Fuel Performance Modeling of U-Mo Monolithic Fuel

As the physics that governs the microstructural evolution of nuclear fuel span various time and spatial scales, to fully understand the fuel behavior inevitably involves atomic to mesoscale resolution that can be difficult to determine experimentally. Microstructural-level modeling and simulations can be used to develop physics-based materials models that can provide physical understanding to inform fabrication process control, as well as a valuable feedback mechanism between post-irradiation examination (PIE) results and fabrication parameters. In accordance with the program schedule, the primary goals of the microstructure modeling effort are to: 1. Address critical microstructural questions and provide practical guidance to the fabricator via the fuel product specification; 2. Provide mechanistic inputs for the existing fuel performance code to improve its descriptive and predictive capability at the macroscopic scale. In fiscal year (FY)-21, the work scope consisted of six main facets: (1) the effect of carbides on fuel performance; (2) gas diffusivity in different phases; (3) integration of microstructural fuel performance modeling; (4) property degradation; (5) irradiation creep; and (6) historical analysis of microstructure data. Brief summaries of each are included below.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

DOE OSTI2020

PREEMPT: Scalable Epidemic Interventions Using Submodular Optimization on Multi-GPU Systems

Preventing and slowing the spread of epidemics is achieved through techniques such as vaccination and social distancing. Given practical limitations on the number of vaccines and cost of administration, optimization becomes a necessity. Previous approaches using mathematical programming methods have shown to be effective but are limited by computational costs. In this work, we make several contributions: First, we present a new approach for intervention via maximizing the influence of vaccinated nodes on the network. We call this method \preempt. Next, we prove submodular properties associated with the objective function of our method so that it aids in construction of an efficient greedy approximation strategy. Consequently, we present a new parallel algorithm based on greedy hill climbing for \preempt, and present an efficient parallel implementation for distributed CPU-GPU heterogeneous platforms. Our results demonstrate that \preempt{} is able to achieve a significant reduction (up to 6.75$\times$) in the percentage of people infected on a city-scale network. We also show strong scaling results of \preempt{} on 128 nodes of the Summit supercomputer. Our parallel implementation is able to significantly reduce time to solution, from hours to minutes on large networks. This work represents a first-of-its-kind effort in parallelizing greedy hill climbing and applying it toward devising effective interventions for epidemics.

Minutoli, Marco↗

DOE OSTI2020

Performance Potential of Mixed Data Management Modes for Heterogeneous Memory Systems

Many high-performance systems now include different types of memory devices within the same compute platform to meet strict performance and cost constraints. Such heterogeneous memory systems often include an upper-level tier with better performance, but limited capacity, and lower-level tiers with higher capacity, but less bandwidth and longer latencies for reads and writes. To utilize the different memory layers efficiently, current systems rely on hardware-directed, memory -side caching or they provide facilities in the operating system (OS) that allow applications to make their own data-tier assignments. Since these data management options each come with their own set of trade-offs, many systems also include mixed data management configurations that allow applications to employ hardware- and software-directed management simultaneously, but for different portions of their address space. Despite the opportunity to address limitations of stand-alone data management options, such mixed management modes are under-utilized in practice, and have not been evaluated in prior studies of complex memory hardware. In this work, we develop custom program profiling, configurations, and policies to study the potential of mixed data management modes to outperform hardware- or software-based management schemes alone. Our experiments, conducted on an Intel ® Knights Landing platform with high-bandwidth memory, demonstrate that the mixed data management mode achieves the same or better performance than the best stand-alone option for five memory intensive benchmark applications (run separately and in isolation), resulting in an average speedup compared to the best stand-alone policy of over 10 %, on average.

Effler, Chad↗