SEARCH · Engineering Papers

Results for “performance based design”

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 721 records · Page 40

All sources NASA NTRS DOE OSTI

DOE OSTI2023

Spread Spectrum Time Domain Reflectometry (SSTDR) and Frequency Domain Reflectometry (FDR) for Detection of Cable Anomalies Using Machine Learning

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2024

SSTDR and FDR Detection of Un-Energized and Energized Cable Anomalies Including Thermal Degradation Using Machine Learning

Historically, cables are initially qualified for nuclear power plant use for 40 years. As plants extend their operating license to 60 and 80 years, continued use of these cables must shift to a performance-based approach since it is cost prohibitive to completely replace cables that are likely still capable of performing their design function. A variety of cable tests are available and are commonly applied during outages when the cables can be taken out of service. Frequency domain reflectometry (FDR) is one of these test methods that is being more broadly accepted and used because it not only detects anomalies along the cable with a low-voltage signal that does not stress the cable insulation, but the technique also locates the anomalies. This supports follow-up local inspection and local repair or partial replacement of a damaged cable segment. Currently, FDR testing is only applied to cables that are taken out of service since the test instrument would be damaged by operational voltages.

36 MATERIALS SCIENCE↗

DOE OSTI2023

Spread Spectrum Time Domain Reflectometry (SSTDR) and Frequency Domain Reflectometry (FDR) for Detection of Cable Anomalies Using Machine Learning

Cables are initially qualified for nuclear power plant use for 40 years. As plants extend their operating license to 60 and 80 years, continued use of these cables must shift to a performance-based approach since it is cost prohibitive to completely replace cables that are likely still capable of performing their design function. A variety of cable tests are available and are commonly applied during outages when the cables can be taken out of service. Frequency domain reflectometry (FDR) is one of these test methods that is being more broadly accepted and used because it not only detects anomalies along the cable with a low-voltage signal that does not stress the cable insulation, but the technique also locates the anomalies. This supports follow-up local inspection and local repair or partial replacement of a damaged cable segment. Currently, FDR testing is only applied to cables that are taken out of service since the test instrument would be damaged by operational voltages. A related technology that has found some acceptance in the aircraft and rail industry is spread spectrum time domain reflectometry (SSTDR). This technology has been implemented with a custom commercial instrument by LiveWire Innovation Inc. that is designed to operate on live cables up to 1000 volts. One of the main conclusions of a previous effort was that cable reflectometry plots can be difficult for humans to analyze due to baseline noise, low or noisy anomaly response peaks, or large responses from cable ends. Detection of cable anomalies for many of these frequencies and test conditions was challenging for manual analysis. This presented an ideal opportunity for ML analysis to distinguish undamaged cable indications from anomalous cable indications. This research discusses application of machine learning (ML) to reflectometry cable test methods. The goal was to assess feasibility to distinguish undamaged cable reflectometry responses from damaged or anomalous cable reflectometry responses. The assessment considered the 3 instruments, multiple frequency bandwidths from each instrument, multiple cable anomalies and test conditions, and both supervised and unsupervised ML approaches. Although approaches and analysis methods were not identical or directly comparable, both outputs were encouraging. The unsupervised prediction weighted accuracy was assessed by instrument and by frequency. It performed better at high frequencies with the highest prediction accuracy of 0.84 for the higher frequency FDR, 0.79 for the 48-MHz LiveWire SSTDR, and 0.77 for 300-MHz PNNL SSTDR. The initial weighted accuracy average across all frequencies for using supervised ML was 0.56 to 0.68. The supervised analysis was repeated with noisier training data removed resulting in weighted accuracies of 0.69 to 0.87. These weighted accuracies are not directly comparable due to differences in the supervised and unsupervised analysis details but do indicate an encouraging trend. Even with limited and unbalanced data, strong prediction accuracies seem encouraging for further work including more data under a wider range of conditions.

42 ENGINEERING↗

DOE OSTI2023

Ionizing Radiation Sensing with Functionalized and Copper-Coated SWCNT/PMMA Thin Film Nanocomposites

This paper studies the ionizing radiation effects on functionalized single-walled carbon nanotube (SWCNT)/poly(methyl methacrylate) (PMMA) thin-film nanocomposites [SWNT/PMMA]. The functionalized thin-film devices are made of ferrocene-doped SWCNTs, SWCNTs functionalized with carboxylic acid (COOH), and SWCNTs coated/ modified with copper. The nanocomposite was synthesized by the solution blending method and the resulting nanocomposite was spin-cast on interdigitated electrodes (IDEs). A 160 kV X-ray source was used to irradiate the thin film and changes in the electrical resistance of the nanocomposites due to X-rays were measured using a semiconductor device analyzer. Carboxylic acid functionalized and copper-coated SWCNT/PMMA nanocomposite showed a reduced response to X-rays compared to unfunctionalized SWCNT/PMMA nanocomposite. Ferrocene-doped SWCNT showed a higher sensitivity to X-rays at lower dose rates. This is in contrast to a previous study that showed that similar nanocomposites using functionalized multi-walled CNTs (MWCNTs) had demonstrated an improved response to X-rays ionizing radiation compared to unfunctionalized MWCNTs for all dose rates. Electrical measurements were also performed using the Arduino Nano microcontroller. The result showed that a relatively economical, lightweight-designed prototype radiation sensor based on SWCNT/PMMA thin-film devices could be produced by interfacing the devices with a modest microcontroller. This work also shows that by encapsulating the SWCNT/PMMA thin-film device in a plastic container, the effect of ambient humidity can be reduced and the device can still be used to detect X-ray radiation. This study further shows that the sensitivity of SWCNT to X-rays was dependent on both the functionalization of the SWCNT and the dose rate.

36 MATERIALS SCIENCE↗

DOE OSTI2020

Prognosis of Wind Turbine Gearbox Bearing Failures Using SCADA and Modeled Data

Predictive maintenance and condition monitoring systems for wind turbines have seen increased adoption to minimize downtime, reducing operation and maintenance costs. On today’s wind power plants, the integrated supervisory control and data acquisition (SCADA) system provides low- frequency operational data that can be leveraged to quantify a wind turbine’s health. The aim of this study is to utilize machine-learning techniques to predict axial cracking failures in wind turbine gearbox bearings up to 1 month ahead of time. The failures are assumed to have occurred when the investigated bearing was replaced. While current SCADA systems show the overall condition of a wind turbine, often they do not allow for the investigation of specific gearbox bearings’ health. To enrich bearing fault signatures, additional data are computed through physics-based models using gearbox design information. Based on SCADA data, modeled data, and bearing failure log data from an actual wind plant, the performances of different machine-learning models on unseen data are then evaluated using industry-standard metrics such as precision, recall, and F1 score. Results show the overall system performance enhancement in predicting bearing failure when modeled data are included with SCADA data. The reduction in terms of false alarms is about 50%, and improvement in terms of precision and F1 score is about 33% and 12% respectively, based on the best modeling case in this study.

49 EE - Wind and Water Power Program - Wind (EE-4W↗

DOE OSTI2020

A Numerical Study on the Energy Performance of a Novel Furnace With Acidic Gas Trap Absorbers

Natural gas furnaces are widely used in US residential and commercial building markets. An important issue for natural gas furnaces is serious corrosion and fouling problems caused by acidic gas, such as SOx. An advanced adsorption technology based on acidic gas trap (AGT) absorbers offers the possibility to remove SOx acidic gas from natural gas furnaces with high efficiency and low cost, thereby enabling the development of condensing furnaces without the use of expensive corrosion resistant materials in the heat exchanger. A three-dimensional (3D) computational fluid dynamics (CFD) model has been developed to evaluate the heat transfer performance of a furnace with AGT absorbers and to compare it with a baseline conventional furnace without the AGT. Moreover, an axisymmetric model has been built focusing on the absorbing process in the AGT. The baseline conventional furnace used for the study is a commercial condensing furnace (Rheem 92% AFUE 84,000 BTU Multi-Position Gas Furnace). This furnace was completely disassembled, and the dimensions of each part were carefully measured and used to build a detailed CFD model. A model representing the new furnace, incorporating the AGT absorbers, was developed by adding the AGT system to the conventional furnace model. For the CFD analysis, a mixture model was employed to characterize the heat and mass transfer during the condensing process in the furnace while considering three components—air, water vapor and liquid water. Condensation takes place in the condensing heat exchanger, where water vapor changes phase to liquid water, and the latent heat is thus used in the furnace for useful heating. The simulation results characterize the energy performance of both the conventional furnace and the novel furnace with AGT absorbers, as well as the reactive processing in the AGT. These results provide insightful guidance for the development of the AGT absorber-based furnace from the perspective of its energy performance and will be used to further optimize this novel furnace design.

Laclair, Tim↗

DOE OSTI2026

Data Center High-Temperature Liquid Cooling and Heat Reuse Techno-Economic Study: Preprint

Data centers are energy-intensive facilities with growing demands for efficiency and cost-effective operations. Smaller, more distributed edge inference data centers are expected to proliferate as AI applications require low latency closer to the user of AI tools, which presents a growing opportunity to explore the systems implications of liquid cooling on water and energy use. This study analyzes the implementation of high-temperature liquid cooling systems in a prototypical inference 1-MW data center and explores the potential for heat reuse across varying climates with a goal to optimize energy efficiency, reduce capital and operational costs, and identify opportunities for high-performance cooling and water use reduction infrastructure. This analysis evaluated configurations utilizing a peak day hourly sizing and systems performance spreadsheet to evaluate design and operational conditions from which component sizes, installed cost, operational cost, and performance metrics were determined for the Base case and the Elevated case. The techno-economic analysis included heat reuse applications across a range of heat recovery temperatures and heat rejection options. The analysis shows that high-temperature liquid cooling allows for improved energy efficiency, lower water consumption, and lower capital costs compared to traditional cooling approaches. Transitioning to elevated water inlet/outlet temperatures (50 degrees C/60 degrees C) eliminates the need for chillers, cooling towers, and heat recovery equipment in many scenarios across three distinct climate zones. This results in up to 75% capital cost savings for the cooling and heat recovery equipment, and with significantly reduced water consumption, especially in non-heat reuse applications. Heat generated from data centers can also be repurposed for space heating, domestic hot water, and other applications, and is most cost-effective when data center outlet temperatures exceed 55-60 degrees C.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2020

Validation of Hygrothermal Simulations with Wall Performance Experiments in an Environmental Chamber

Oak Ridge National Laboratory is developing a web tool, built on a rule-based expert system, that aids stakeholders in designing energy-efficient moisture-durable walls. Currently, the tool’s expert system database is mostly populated with expert opinion, but work is being done to incorporate durability assessment based on stochastic hygrothermal modeling. Chamber experiments have been conducted to validate the hygrothermal models. The measured temperature and humidity have been compared with those predicted by the hygrothermal model. Experiments and comparison to one-dimensional hygrothermal modeling results were completed with two walls, a Structural Insulated Panel (SIP) based wall and a Concrete Masonry Unit (CMU) based wall. The two walls were succumbed to typical Chicago weather in ORNL’s Heat, Air and Moisture chamber. The walls were sequentially exposed to different scenarios, including diffusion, wetting, air leakage and solar radiation. For the most part, the hygrothermal simulations can be successfully used to predict the performance of these wall systems for the moisture transport phenomenon. Errors between measured and simulated values decreased as measurements got closer to the interior side of the wall. The root mean square error was larger for relative humidity (up to 17.5%-RH for CMU) than for temperature (up to 4.5°C for CMU wall). The errors were larger for the CMU wall than the wood frame wall. The phenomenon, including liquid water, caused large discrepancies between measurement and simulation results, and simulated results showed slower drying of materials than measured results. The one-dimensional nature of the simulation model made simulating air leaks difficult but not impossible.

Salonvaara, Mikael↗

DOE OSTI2020

SDU7 Interior Liner Testing & Evaluation

The Saltstone Disposal Unit 7 (SDU7) project requested the Savannah River National Laboratory (SRNL), Materials Science & Engineering (MS&E) organization to evaluate an alternative bromobutyl liner and adhesives for potential use in SDU7 based on applicable ASTM testing standards. SRNL performed similar testing for the liner system used in SDU6. Bonded and non-bonded samples of the alternative liner (Blair Rubber Marflex™ RCHB60HT) were subjected to specific ASTM tests after immersion in two Saltstone leachate simulants, designated S1 and S2, for 1000 hours at 60 °C. Immersion exposures were performed at the Savannah River Ecology Laboratory (SREL). Post-immersion testing involved mechanical property and hardness testing of base material, lap-shear testing of bonded samples and rubber-to-concrete paver interrogation. The liner exhibited an approximate 40% drop in tensile strength and 25% drop in elongation at failure from baseline values after immersion, though final values are comparable to previous SDU6 liner values and are within liner manufacturer property ranges. Overall lap-shear strength behavior of the four adhesives showed a similar pattern, with short-term reduction and subsequent increase in peak load values. However, ranking of adhesives varied with the metric used for comparison. Two of the four adhesives tested (Normac 900 and REMA SC 4000) showed overall better behavior, collectively considering immersion performance, lap-shear data and bonded paver interrogation. Lap-shear samples bonded with the REMA adhesive showed the best combination of final retained bond strength, 6-week average bond strength and bond failure mode. Only one paver (REMA SC4000) was noted to have no defects after immersion. This document details the testing performed and provides conclusions and recommendations. The testing suggests that SDU liner performance is highly dependent upon seam integrity, which collectively depends on a combination of adhesive properties, installation workmanship and inspection/quality control.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2021

Learning Sequential Distribution System Restoration via Graph-Reinforcement Learning

We report a distribution service restoration algorithm as a fundamental resilient paradigm for system operators provides an optimally coordinated, resilient solution to enhance the restoration performance. The restoration problem is formulated to coordinate distribution generators and controllable switches optimally. A model-based control scheme is usually designed to solve this problem, relying on a precise model and resulting in low scalability. To tackle these limitations, this work proposes a graph-reinforcement learning framework for the restoration problem. We link the power system topology with a graph convolutional network, which captures the complex mechanism of network restoration in power networks and understands the mutual interactions among controllable devices. Latent features over graphical power networks produced by graph convolutional layers are exploited to learn the control policy for network restoration using deep reinforcement learning. The solution scalability is guaranteed by modeling distributed generators as agents in a multi-agent environment and a proper pre-training paradigm. Comparative studies on IEEE 123-node and 8500-node test systems demonstrate the performance of the proposed solution.

24 POWER TRANSMISSION AND DISTRIBUTION↗

DOE OSTI2025

An optimization approach for cryogenic distillation-based protium removal systems in magnetic fusion energy fuel cycles

Tritium inventory reduction for fusion power plants is central to the successful adoption of fusion energy. The advent of direct internal recycling in deuterium-tritium fusion fuel cycle design has led to significant reduction in startup tritium inventory estimates for fusion power plants but requires an isotope rebalancing and protium removal (IRPR) system to ensure acceptable isotopic fuel composition. Cryogenic distillation is a potential solution for near-term deployment in an IRPR system due to its demonstrated performance in similar operating regimes. Using protium removal in the fuel cycle as the primary performance metric, this paper presents an optimization methodology for a single-column continuous cryogenic distillation-based IRPR system for a 500 MW fus magnetic fusion device. Distillation column optimization was performed using the CryOgenic Distillation For Isotopic Separation of Hydrogen (CODFISH) code developed at Savannah River National Laboratory. The distillation column design presented maximizes direct recycling of hydrogen isotopes from the fusion chamber exhaust to the fueling system while minimizing IRPR system steady-state tritium inventory. The optimized IRPR distillation column presented achieves direct recycling of 60 % of the hydrogen isotopes in the fusion chamber exhaust with an estimated steady-state system tritium inventory <30 g. In conclusion, the optimized IRPR distillation column operation was then used to estimate the design requirements for a detritation column to treat the IRPR system effluent stream.

CryOgenic distillation↗

DOE OSTI2024

Highly Efficient Regeneration Module for Carbon Capture Systems in NGCC Applications

The objective of this project is to design, fabricate, and test a highly efficient regeneration module capable of providing an ultra-lean absorption solution that is required for capturing CO 2 from dilute sources at 95% or better efficiency. By integrating this advanced regenerator module with SRI International’s Mixed Salt Process (MSP) absorption modules, SRI expects to demonstrate significant progress toward a reduction in cost of capture versus the DOE reference natural gas combined cycle (NGCC) plant with carbon capture. SRI designed, built, and tested an advanced stripper to enhance the performance of SRI’s MSP for CO 2 capture – a transformational ammonia-based solvent technology – for natural gas (NG) power sources. The testing of the advanced stripper for MSP was conducted at an SRI site using a simulated flue gas stream equivalent to about 10 kWe. The research work included modeling of the advanced stripper and integrating it with the MSP absorbers; studying the strategies for producing very highly alkaline lean solvent with minimized emissions; operating the stripper with advanced heat integration to improve process efficiencies; and collecting critically important data for a detailed techno-economic analysis (TEA). The project tasks were designed to address concerns relating to scale-up and integration of the technology to NG power plants—more specifically, to maximize the carbon capture efficiency achievable with MSP and identify pathways to achieve higher capture efficiencies and ultimately zero net carbon emissions. SRI teamed up with a process modeling company (OLI Systems), a process and chemical engineering company (Trimeric Corporation), and a cost-sharing commercial partner (Baker-Hughes – a leading multinational company that designs, manufactures, and services transformative energy technologies) to execute the project. The research findings will accelerate the MSP development and pave the way for the technology to reach the DOE’s goal, and ultimately commercialization of the MSP technology for low-cost CO 2 capture from NGCC flue gas and other dilute CO 2 sources.

03 NATURAL GAS↗

DOE OSTI2021

Enhanced Data Efficiency Using Deep Neural Networks and Gaussian Processes for Aerodynamic Design Optimization

Adjoint-based optimization methods are attractive for aerodynamic shape design primarily due to their computational costs being independent of the dimensionality of the input space and their ability to generate high-fidelity gradients that can then be used in a gradient-based optimizer. This makes them very well suited for high-fidelity simulation based aerodynamic shape optimization of highly parametrized geometries such as aircraft wings. However, the development of adjoint-based solvers involve careful mathematical treatment and their implementation require detailed software development. Furthermore, they can become prohibitively expensive when multiple optimization problems are being solved, each requiring multiple restarts to circumvent local optima. In this work, we propose a machine learning enabled, surrogate-based framework that replaces the expensive adjoint solver, without compromising on predicting predictive accuracy. Specifically, we first train a deep neural network (DNN) from training data generated from evaluating the high-fidelity simulation model on a model-agnostic design of experiments on the geometry shape parameters. The optimum shape may then be computed by using a gradient-based optimizer coupled with the trained DNN. Subsequently, we also perform a gradient-free Bayesian optimization, where the trained DNN is used as the prior mean. We observe that the latter framework (DNN-BO) improves upon the DNN-only based optimization strategy for the same computational cost. Overall, this framework predicts the true optimum with very high accuracy, while requiring far fewer high-fidelity function calls compared to the adjoint-based method. Furthermore, we show that multiple optimization problems can be solved with the same machine learning model with high accuracy, to amortize the offline costs associated with constructing our models. Our methodology finds applications in the early stages of aerospace design. (C) 2021 Published by Elsevier Masson SAS.

Renganathan, S. Ashwin↗

DOE OSTI2025

Optimization of an Impedance-Matched Test Fixture with the Modal Projection Error

Across many industries and engineering disciplines, components and systems are designed and deployed into their operational environment of intended use. It is the desire of the design agency to be able to predict whether their component or system will function in its shock and vibration environments or if it will fail due to mechanical stresses. One method to determine if the component will survive the shock and vibration environments is to expose the component to the operational environment in a laboratory. One difficulty in executing a representative laboratory test is that the component may not have the same boundary condition in the laboratory as in the operational configuration. This paper examines the use of parameterized optimization to design the test fixture in order to better match the operational configuration. Several frequency and modal-based objective functions are examined for the optimization. The study shows that the Modal Projection Error objective function performs the best of the functions studied. The efficacy of the Modal Projection Error is demonstrated with respect to dynamic test fixture design on several analytical and experimental exemplars.

dynamic↗

DOE OSTI2022

Regularizing Invertible Neural Networks for Airfoil Design Through Dimension Reduction

This work examines how dimension reduction can improve the performance of invertible neural networks (INN) for airfoil design. Design workflows are typically expensive, relying on many evaluations of high fidelity computational fluid dynamics (CFD) models. Furthermore, the inverse design problem is typically ill-posed. That is, multiple valid solutions exist that satisfy the design criteria. Regularization can reduce this inverse design space and simplify the problem. We study the use of subspace-based input dimension reduction to act as a regularizer for the INN model and improve the recovery of new airfoil shapes with desired performance characteristics. We find that the dimension reduction identifies two dominant modes, relating to airfoil thickness and camber, that optimally determine the airfoil's aerodynamics. We demonstrate the capability of the proposed INN model to generate 100 airfoils that satisfy the specific aerodynamic and structural characteristics.

airfoil design↗

DOE OSTI2021

Machine learning-enabled prediction of chemical durability of A 2 B 2 O 7 pyrochlore and fluorite

Pyrochlore-structure type and its derivative in a general formula A 2 B 2 O 7 (A = rare earth elements and actinides; B = Ti, Sn, Zr, Hf, Pb, Si, etc.) display excellent structural flexibility and rich crystal chemistry as promising nuclear waste form materials capable of immobilizing actinides and fission products. It is essential to understand these materials’ chemical durability and element release of radionuclides in order to evaluate their performance in near-field environment. However, it is a formidable grand technological challenge to experimentally perform durability testing across hundreds of thousands of possibilities resulting from their extreme compositional complexities due to cation substitutions at both A and B-sites. In this work, we demonstrate a machine learning approach to determine the key materials parameters and structural characteristics governing the leaching behaviors from a small set of selected compositions as model systems, enabling a science-based prediction of their chemical durability that can be extended to a wide range of chemical compositions. The combination of four key structural characteristics and materials parameters, including ionic radius size difference , ionic potential difference , electronegativity difference , and lattice parameter , creates features an optimized prediction of the chemical durability. Two machine learning models, linear regression and Kernel ridge regression models, are trained on the randomly-split training dataset derived from the experimentally-determined elemental release rates, and subsequently tested on the testing dataset. The predicted leaching rates from both machine learning models show an excellent agreement with the experimental data, demonstrating the feasibility of rapidly evaluating the material properties of new compositions. These results highlight the immense potential of synergizing informatics through machine learning-based models and well-controlled experiments of selected model systems to accelerate materials design and discovery with optimized compositions and performance of promising materials for effective nuclear waste management.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2021

The Impact of Different Substituents in Fluorinated Cyclic Carbonates in the Performance of High Voltage Lithium-Ion Battery Electrolyte

Ethylene carbonate (EC) has been used as the Solid-Electrolyte Interphase (SEI) former in the conventional electrolyte for decades. However, its low anodic stability leads to severe capacity decay during cycling under high voltage operation. Therefore, finding a viable electrolyte with high anodic stability and the ability to form robust SEI for high voltage lithium-ion batteries is of primary importance. In this study, a series of electrolytes containing various fluorinated cyclic carbonates as the SEI former have been designed, synthesized and evaluated. Linear sweep voltammetry study suggested that fluorinated cyclic carbonates generally possess higher anodic stability than EC. Based on the cycling performance of LiNi 0.5 Co 0.2 Mn 0.3 O 2 (NMC532)/ graphite full cells, the electrolyte with DFEC/FEMC (1.0 M LiPF 6 ) inhibited the oxidation side reaction on the cathode and forms a vigorous SEI on the anode. The high voltage NMC532/Graphite cell utilizing the novel electrolyte exhibits excellent cycling durability under both room and elevated temperature. The superior performance of the DFEC based electrolyte was further unveiled by SEM, EDAX and XRD. Both the anode and cathode of the full cell employing DFEC based electrolyte retained their intrinsic structures after cycling while the electrodes cycled in conventional electrolyte showed severe degradation.

25 ENERGY STORAGE↗

DOE OSTI2021

Co-design of Advanced Architectures for Graph Analytics using Machine Learning

A graph is an excellent way of representing relationships among entities. We can use graph analytics to synthesize and analyze such relational data, and extract relevant features that are useful for various tasks such as machine learning. Considering the crucial role of graph analytics in various domains, it is important and timely to investigate the right hardware configurations that can achieve optimal performance for graph workloads on future high-performance computing systems. Design space exploration studies facilitate the selection of appropriate configurations (e.g. memory) to achieve a desired system performance. Recently, the approach of accelerating graph analytics using persistent non-volatile memory has gained a lot of attention. Traditional system simulators such as Gem5 and NVMain can be used to explore the design space of these advanced memory architectures for graph workloads. However, these simulators are slow in execution thus limiting the efficiency of design space exploration studies. To overcome this challenge, we proposed a machine learning based approach to co-design advanced memory architectures for graph workloads. We tested our approach with DRAM, non-volatile memory, and hybrid memory (DRAM+NVM) using a breadth first search benchmark algorithm. Our results showed the applicability of the proposed machine learning based approach to the co-design of the advanced memory architectures. In this paper, we provide recommendations on selecting advanced memory architectures to achieve desired performance for graph workloads. We also discuss the performances of different machine learning models that were considered in this study.

Kurte, Kuldeep↗