Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “simulation decomposition”

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 181 records · Page 10
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Rendezvous algorithms for large-scale modeling and simulation

Rendezvous algorithms encode a communication pattern that is useful when processors sending data do not know who the receiving processors should be, or vice versa. The idea is to define an intermediate decomposition where datums from different sending processors can ”rendezvous” to perform a computation, in a manner that both the senders and eventual receivers of the results can identify the appropriate rendezvous processor. Though they were originally designed for interpolating between overlaid grids with independent parallel decompositions (Plimpton et al., 2004), we have recently found rendezvous algorithms useful for a variety of operations in particle- or grid-based simulation codes when running large problems on large numbers of processors. In particular, we show they can perform well when a load-balanced intermediate decomposition is randomized and not spatial, requiring all-to-all communication to move data between processors. In this case rendezvous algorithms leverage the large bisection communication bandwidths which parallel machines provide. We describe how rendezvous algorithms work in a scientific computing context and give specific examples for molecular dynamics and Direct Simulation Monte Carlo codes which result in dramatic performance improvements versus simpler algorithms which do not scale as well. We explain how a generic rendezvous algorithm can be implemented, and also point out similarities with the MapReduce paradigm popularized by Google and Hadoop.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Pressure–strain interaction. III. Particle-in-cell simulations of magnetic reconnection

How energy is converted into thermal energy in weakly collisional and collisionless plasma processes, such as magnetic reconnection and plasma turbulence, has recently been the subject of intense scrutiny. The pressure–strain interaction has emerged as an important piece, as it describes the rate of conversion between bulk flow and thermal energy density. In two companion studies, we presented an alternate decomposition of the pressure–strain interaction to isolate the effects of converging/diverging flow and flow shear instead of compressible and incompressible flow, and we derived the pressure–strain interaction in magnetic field-aligned coordinates. Here, we use these results to study pressure–strain interaction during two-dimensional anti-parallel magnetic reconnection. We perform particle-in-cell simulations and plot the decompositions in both Cartesian and magnetic field-aligned coordinates. We identify the mechanisms contributing to positive and negative pressure–strain interaction during reconnection. Furthermore, this study provides a roadmap for interpreting numerical and observational data of the pressure–strain interaction, which should be important for studies of reconnection, turbulence, and collisionless shocks.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2026

Local reduced-order modeling for electrostatic plasmas by physics-informed solution manifold decomposition

Despite advancements in high-performance computing and modern numerical algorithms, computational cost remains prohibitive for multi-query kinetic plasma simulations. Here, in this work, we develop data-driven reduced-order models (ROMs) for collisionless electrostatic plasma dynamics, based on the kinetic Vlasov-Poisson equation. Our ROM approach projects the equation onto a linear subspace defined by the proper orthogonal decomposition (POD) modes. We introduce an efficient tensorial method to update the nonlinear term using a precomputed third-order tensor. We capture multiscale behavior with a minimal number of POD modes by decomposing the solution manifold into multiple time windows and creating temporally local ROMs. We consider two strategies for decomposition: one based on the physical time and the other based on the electric field energy. Applied to the 1D1V Vlasov–Poisson simulations, that is, prescribed E-field, Landau damping, and two-stream instability, we demonstrate that our ROMs accurately capture the total energy of the system both for parametric and time extrapolation cases. The temporally local ROMs are more efficient and accurate than the single ROM. In addition, in the two-stream instability case, we show that the energy-windowing reduced-order model (EW-ROM) is more efficient and accurate than the time-windowing reduced-order model (TW-ROM). With the tensorial approach, EW-ROM solves the equation approximately 90 times faster than Eulerian simulations while maintaining a maximum relative error of 7.5% for the training data and 11% for the testing data.

Electrostatic plasmas↗
DOE OSTI2023

Trapped-ion quantum simulation of collective neutrino oscillations

It is well known that the neutrino flavor in extreme astrophysical environments changes under the effect of three contributions: the vacuum oscillation, the interaction with the surrounding matter, and the collective oscillations due to interactions between different neutrinos. The latter adds a nonlinear contribution to the equations of motion, making the description of their dynamics complex. In this work we study various strategies to simulate the coherent collective oscillations of a system of N neutrinos in the two-flavor approximation using quantum computation. This was achieved by using a pair-neutrino decomposition designed to account for the fact that the flavor Hamiltonian, in the presence of the neutrino-neutrino term, presents an all-to-all interaction that makes the implementation of the evolution dependent on the qubit topology. We analyze the Trotter error caused by the decomposition demonstrating that the complexity of the implementation of time evolution scales polynomially with the number of neutrinos and that the noise from near-term quantum device simulation can be reduced by optimizing the quantum circuit decomposition and exploiting a full-qubit connectivity. We find that the gate complexity using second order Trotter-Suzuki formulas scales better with system size than with other decomposition methods such as quantum signal processing. In conclusion, we finally present the application and the results of our algorithm on a real quantum device based on trapped-ion qubits.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

BlueCRAB Domain Overlapping Coupling: Theory & Verification

Coupling low- and high-fidelity codes is a useful way to model complex engineering systems. Thanks to the high-fidelity code, complex phenomena can be resolved in areas of the system where this is required, yet the efficiency of the lower-fidelity code is still retained in modeling the rest of the system. This document details the theory and implementation of coupling two different thermal-hydraulic codes: the system thermal-hydraulics (STH) code System Analysis Module (SAM) and the coarse-mesh computational fluid dynamics (CFD) code Pronghorn. Both applications are included in the Comprehensive Reactor Analysis Bundle (BlueCRAB) code suite and are based on the Multiphysics Object-Oriented Simulation Environment (MOOSE) framework. The domain overlapping (DO) coupling approach was adopted, as it offers proven advantages over more conventional domain decomposition methods. In the DO coupling, SAM provides Pronghorn with boundary conditions that depend on the system-level simulation of the entire plant. In return, the overlapping coupled SAM components–termed “surrogate components”–receive friction factors and source terms computed online based on the Pronghorn simulation. The framework is designed to be generic and enable coupling regardless of geometry and the number of inlet/outlet boundaries in the DO coupled domain. The developed method leads to consistent pressure drops, enthalpies, and scalar concentrations when comparing the coupled SAM and Pronghorn simulations. This document presents the DO coupling approach, along with two verification and two demonstration cases. The proposed problems explore different physical aspects relevant to nuclear reactor analysis, including buoyancy-driven flows, complex flow patterns, and multiple inlets and outlets. Periodically, new versions of this “BlueCRAB Domain Overlapping Coupling: Theory & Verification” report will be issued to reflect future developments and verification tests.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2022

Dynamic mode decomposition with core sketch

With the increase in collected data volumes, either from experimental measurements or high fidelity simulations, there is an ever-growing need to develop computationally efficient tools to process, analyze, and interpret these datasets. Modal analysis techniques have gained great interest due to their ability to identify patterns in the data and extract valuable information about the system being considered. Dynamic mode decomposition (DMD) relies on elements of the Koopman approximation theory to compute a set of modes, each associated with a fixed oscillation frequency and a decay/growth rate. Extracting these details from large datasets can be computationally expensive due to the need to implement singular value decomposition of the input data matrix. Sketching algorithms have become popular in numerical linear algebra where statistical theoretic approaches are utilized to reduce the cost of major operations. A sketch of a matrix is another matrix, which is significantly smaller, but still sufficiently approximates the original system. We put forth an efficient DMD framework, SketchyDMD, based on a core sketching algorithm that captures information about the range and corange (their mutual relationship) of input data. The proposed sketching-based framework can accelerate various portions of the DMD routines, compared to classical methods that operate directly on the raw input data. We conduct numerical experiments using the spherical shallow water equations as a prototypical model in the context of geophysical flows. In conclusion, we show that the proposed SketchyDMD is superior to existing randomized DMD methods that are based on capturing only the range of the input data.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2022

Numerical investigation of the influence of shear and thermal stratification on the wind turbine tip‐vortex stability

Summary The interaction between wind turbine wakes and atmospheric turbulence is characterised by complex dynamics. In this study, two major components of the atmospheric boundary layer dynamics have been isolated, namely, the mean velocity profile shear and the thermal stratification, to examine their impact on the near‐wake development by undertaking a series of highly resolved large‐eddy simulations. Subsequently, instantaneous flow fields are extracted from the simulations and used to conduct Fourier analysis and proper orthogonal decomposition (POD) and compute the mean kinetic energy fluxes by different POD modes to better understand the tip‐vortex instability mechanisms. Our findings indicate that shear can significantly affect the breakup of the wind turbine tip‐vortices and the shape and stable length of the wake, whereas thermal stratification seems to only have limited contribution to the spatial development of the near‐wake field. Finally, our analysis shows that the applied perturbation frequency determines the tip‐vortex breakup location as it controls the onset of the mutual inductance instability.

17 WIND ENERGY↗
DOE OSTI2024

Statistical data analysis of x-ray spectroscopy data enabled by neural network accelerated Bayesian inference

Bayesian inference applied to x-ray spectroscopy data analysis enables uncertainty quantification necessary to rigorously test theoretical models. However, when comparing to data, detailed atomic physics and radiation transfer calculations of x-ray emission from non-uniform plasma conditions are typically too slow to be performed in line with statistical sampling methods, such as Markov Chain Monte Carlo sampling. Furthermore, differences in transition energies and x-ray opacities often make direct comparisons between simulated and measured spectra unreliable. Here, we present a spectral decomposition method that allows for corrections to line positions and bound–bound opacities to best fit experimental data, with the goal of providing quantitative feedback to improve the underlying theoretical models and guide future experiments. In this work, we use a neural network (NN) surrogate model to replace spectral calculations of isobaric hot-spots created in Kr-doped implosions at the National Ignition Facility. The NN was trained on calculations of x-ray spectra using an isobaric hot-spot model post-processed with Cretin, a multi-species atomic kinetics and radiation code. The speedup provided by the NN model to generate x-ray emission spectra enables statistical analysis of parameterized models with sufficient detail to accurately represent the physical system and extract the plasma parameters of interest.

47 OTHER INSTRUMENTATION↗
DOE OSTI2024

Chasing the multi-modal plasma response in MAST-U

Abstract Achieving edge localized modes (ELMs) suppression in spherical tokamaks by applying resonant magnetic perturbations (RMPs) has proven challenging. The poloidal spectrum of the applied RMP is a key parameter that has an impact on the capability to mitigate and eventually suppress ELMs. In this work the resistive magnetohydrodynamic code MARS-F (Liu et al 2000 Phys. Plasmas 7 3681) is used to evaluate the possibility of directly measuring the plasma response in MAST-U, and particularly its variation as function of the applied poloidal spectrum, in order to guide the experimental validation of the predicted best RMP configuration for ELM suppression. Toroidal mode number n = 2 RMP is considered to minimize the presence of sidebands, and to avoid the deleterious core coupling of n = 1. Singular Value Decomposition is used to highlight linearly independent structures in the simulated magnetic 3D fields and how those structures can be measured at the wall where the magnetic sensors are located. Alternative ways to measure the multimodal plasma response and how they can be used to infer the best RMP configuration to achieve ELM suppression are also presented, including the plasma displacement and the 3D footprints at the divertor plates.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2024

Extracting the partonic structure of colorless exchanges at the Electron Ion Collider

We investigate the determination of the partonic structure of colorless exchanges in deep inelastic diffractive e p scattering at the Electron Ion Collider (EIC), using the standard decomposition into Pomeron and Reggeon contributions. We perform fits to simulated diffractive cross section pseudodata in four variables, including the momentum transfer t , to estimate the achievable precision on the Pomeron and Reggeon quark and gluon distributions. We analyze the influence of different cuts in the kinematic variables, beam energy configurations, and luminosities, including a “first year” scenario. We conclude that the EIC will be able to constrain the partonic structure of the subleading Reggeon exchange with a precision comparable to that of the leading Pomeron exchange. Published by the American Physical Society 2024

Astronomy & Astrophysics↗
DOE OSTI2021

A Functional All-Hazard Approach to Critical Infrastructure Dependency Analysis

The critical infrastructures protection landscape is a vast and varied pattern of independent, but interconnected infrastructure systems that are essential to the function of our modern society. The U.S. policy on critical infrastructure protection has been continually evolving since the “President’s Commission on Critical Infrastructure Protection” was published in 1997. In response to these policies, federal, state, and local governments, along with research institutions, have invested a substantial amount of time and effort into identifying and analyzing critical infrastructure, their functions, and dependencies/interdependencies to better understand their vulnerabilities. To date, the ability to assess vulnerabilities, resiliency, and priorities for protecting interdependent critical infrastructure systems from an all-hazards perspective remains a difficult problem. In this paper we introduce the All-Hazards Analysis (AHA) methodology, which provides an integrated functional basis across infrastructure systems, through the implementation of a common language and a scalable level of decomposition to effectively evaluate the resilience of interconnected infrastructure systems. AHA models infrastructure systems as directed multidimensional graphs, which enable the evaluation of cross-sector interdependencies prior to, during, and after disruptive events. Finally, and by design, AHA enables the cross linking of data taxonomies to enable more effective data sharing, such as the National Critical Functions (NCF) and Infrastructure Data Taxonomy (IDT).

02 PETROLEUM↗
DOE OSTI2024

Comparison of volatiles evolving from selected highland and mare lunar regolith simulants during vacuum sintering

Volatiles evolving from JSC-1A, NU-LHT-4M and CSM-LHT-1G lunar regolith simulants during in vacuo thermal processing were analyzed using mass spectrometry as a function of temperature. Two high-fidelity simulants, JSC-1A (mare) and NU-LHT-4M (highland), were compared to a newly developed CSM-LHT-1G highland simulant, modified to closely match lunar geochemistry. Large autogenous gas loads were observed for all investigated materials. Mineralogical knowledge was used to identify and attribute individual volatile species to reacting, transforming, or decomposing constituents (hydrates, carbonates, sulfates, sulfides, clays, etc.) of the respective regolith simulant in the self-generated gas environment. Cumulative mass losses for individual simulant components as a function of temperature were quantified using mass spectrometry in conjunction with thermogravimetric analysis. Investigation of the four components of CSM-LHT-1G – anorthosite, basalt, augite, and glass – aided the attribution of volatile species to specific compounds and their respective sources. The results showed significant decomposition of non-lunar phases present in the man-made regolith simulants below the typical glass crystallization temperatures, which paves the way to devising methods for enhancing the fidelity of the simulants. Finally, high gas loads and corrosive gases (HF and HCl) were recognized as potential hazards, pertaining to the development of large testbed facilities.

42 ENGINEERING↗
DOE OSTI2026

pyRMG: A framework for high-throughput, large-cell DFT calculations on supercomputers

Exascale computing delivers the raw power to simulate ever larger and more chemically realistic systems, but realizing this potential requires codes that can efficiently use thousands of processors. Our real-space multigrid (RMG) density functional theory (DFT) code’s grid-decomposition approach scales nearly linearly with the number of graphics processing units (GPUs), even for simulations exceeding thousands of atoms. This scalability makes RMG a compelling tool for high-throughput DFT studies of materials that would otherwise be bottlenecked in other codes (for example, by global fast Fourier transforms in plane-wave DFT). However, the limited workflow infrastructure for RMG has thus far constrained its adoption to a small user community. In this work, we present pyRMG, a Python package designed to streamline the setup and execution of RMG DFT calculations. Built on the pymatgen and ASE (Atomic Simulation Environment) computational materials science Python packages, pyRMG automates input generation and convergence checking, and it integrates with modern job schedulers (e.g., Flux) on leadership-class platforms such as Frontier and Perlmutter. Here, we demonstrate pyRMG for a high-throughput study of strain effects in 2D 2L-Bi 2 Se 3 /2L-NbSe 2 heterostructures, which offers chemical insights into this system and shows that RMG-based workflows can converge with limited user intervention.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Pit rim decomposition into multiple quantum dots on surfaces of epitaxial thin films grown on pit-patterned substrates

Here, we report results of dynamical simulations according to an experimentally validated surface morphological evolution model on the formation of regular arrays of quantum dot molecules (QDMs) consisting of 1D arrays of smaller interacting quantum dots (QDs). These QD arrays form along the sides of each pit rim on the surface of a coherently strained thin film epitaxially deposited on a semiconductor substrate, the surface of which consists of a periodic pattern of inverted prismatic pits with rectangular pit openings. We find that this complex QDM pattern results from the decomposition of the pit rim from a “quantum fortress” with four elongated QDs into four 1D arrays of multiple smaller QDs arranged along each side of the pit rim. Systematic parametric analysis indicates that varying the pit opening dimensions and the pit wall inclination directly impacts the number of QDs in the resulting QDM pattern, while varying the pit depth only affects the dimensions of the QDs in the nanostructure pattern. Therefore, the number, arrangement, and sizes of QDs in the resulting pattern of QDMs on the film surface can be engineered precisely by proper tuning of the pit design parameters. Our simulation results are supported by predictions of morphological stability analysis, which explains the pit rim decomposition into multiple QDs as the outcome of a tip-splitting instability and provides a fundamental characterization of the post-instability nanostructure pattern. Our theoretical findings can play a vital role in designing optimal semiconductor surface patterns toward enabling future nanofabrication technologies.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Surface Crust Formation Controls Evaporation Kinetics of Secondary Organic Aerosols

Gas-particle partitioning is critical for the evolution of secondary organic aerosols (SOA) in the atmosphere. SOA particles evaporate more slowly than expected at nearly size-independent rates, but the underlying mechanism remains controversial. Here, in this study, we apply kinetic multilayer modeling to simulate evaporation of α-pinene SOA, demonstrating that surface crust formation, emerging from accumulation of low-volatility compounds at the particle surface, leads to slow evaporation and reduced size dependence of the evaporation rate. While evaporation induced by decomposition of oligomers would naturally lead to size-independent evaporation rates, we observe and simulate nearly size-independent slow evaporation of polyethylene glycol mixture particles containing polymeric species that do not decompose, confirming the relevance of composition-dependent diffusivity for size-independent, slow evaporation. Slow evaporation of limonene SOA was also observed in environmental chamber experiments, and model simulations demonstrate strong surface crust formation with bulk diffusivity being depressed by up to 5 orders of magnitude compared to the inner bulk. We present experimental evidence using a surface-based mass spectrometry technique that shows that the particle surface becomes enriched in high molecular weight compounds upon evaporation of monomers. Our findings imply that viscous surface crusts may also limit the growth and chemical transformation of SOA particles, influencing their impacts on air quality and climate.

heterogeneous chemistry↗
DOE OSTI2023

Ectomycorrhizal effects on decomposition are highly dependent on fungal traits, climate, and litter properties: A model-based assessment. Dataset.

To simulate the effects of mycorrhizal fungi on soil organic matter cycling, we incorporated mycorrhizal processes into the Carbon, Organisms, Rhizosphere, and Protection in the Soil Environment (CORPSE) model to develop a new soil model Myco-CORPSE. The new model was calibrated and evaluated against soil measurements taken at temperate forests in New Hampshire (NH) and Georgia (GA). A series of scenario analysis were also conducted to explore the conditions under which ectomycorrhizal (ECM) N acquisition processes can induce different soil C accumulation in ECM systems compared to arbuscular (AM) systems.In this data package, we included:-The Python codes of the standard Myco-CORPSE model we developed: "Standard Myco_CORPSE python codes.zip". The main program is the "gradient_sim.py" which calculates the bulk soil microbes and CN content along a user defined gradient of clay, soil temperature, soil moisture and mycorrhizal dominance, and relies on two subprograms "CORPSE_deriv.py" and "CORPSE_integrate.py". "CORPSE_deriv.py" calculated the changes in all simulated soil stock within every time step and "CORPSE_integrate.py" integrate the changes in all simulated soil stock within simulated time period. The program "Plot.py" is used to plot the major outputs produced by the main program "gradient_sim.py".-The modified Python codes of Myco-CORPSE models with site-level environmental inputs (in NH and GA) used to conduct simulations in NH and GA sites: "NH_GA model simulations.zip". -The Python codes used to evaluate the Myco-CORPSE simulation outputs in NH and GA sites against site-level measurements: "Plot NH_GA simulation against measurements.zip". It includes both the evaluation Python code, the model outputs on NH and GA sites, and the measured soil properties in both sites.-The modified Python codes of Myco-CORPSE models "Scenario analysis_model simulations.zip" that is used to conduct scenario analysis of how different litter properties, mycorrhizal fungal traits, climate, and seasonal variation in temperature and vegetation phenology impact the mycorrhizal effects on soil CN properties. The sub file folder "Scenario analysis_litter traits" contains the codes for scenario analysis of different litter properties; The sub file folder "Scenario analysis_ECM types" contains the codes for scenario analysis of different ECM fungal traits; The sub file folder "Scenario analysis_climate&seasonality" contains the codes for scenario analysis of different climate and seasonalities;-"Scenario analysis_model results and plotting codes.zip" contains all the output files from the the scenario analysis of Myco-CORPSE model as described above and the plotting codes used to the generate the heatmaps and scatterplots shown in the manuscript "Ectomycorrhizal effects on decomposition are highly dependent on fungal traits, climate, and litter properties: A model-based assessment"The majority of the model outputs did not have specific geographic information or temporal coverage because the analysis we conducted are mainly hypothetical model simulations. We only provided geographic description, coordinates and temporal coverage for those soil measurements which we used for model evaluations (included in the "Plot NH_GA simulation against measurements.zip").

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Alcohol-Induced Low-Temperature Blockage of Supported-Metal Catalysts for Enhanced Catalysis

The partial or complete blockage of active sites of metal nanoparticles (NPs) on supported-metal catalysts has been of interest for tuning the stability, selectivity, and rate of reactions. In this study, we show that Au-sites in Au/TiO 2 surprisingly become blocked upon treatment in common alcohols (2-propanol and methanol), with 2-propanol causing a greater extent of blockage. Nearly 95% of Au-sites are covered after treatment in 2-propanol at room temperature, followed by desorption at 150 °C. Infrared spectroscopy of CO adsorption unambiguously confirms the occurrence of this phenomenon. Electron energy loss spectroscopy (EELS), temperature-programmed desorption (TPD), Raman spectroscopy, and DFT simulations suggest that the formation of carbon deposits from 2-propanol decomposition and/or the migration of a TiO x layer over the supported NPs may be responsible for the blockage of Au-sites. Nearly full coverage of Au NPs after treatment in 2-propanol led to negligible activity for catalytic CO oxidation, whereas partial retraction of the overlayer led to enhanced activity with time-on-stream, suggesting a self-activating catalytic performance.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Improving Progressive Retrieval for HPC Scientific Data using Deep Neural Network

As the disparity between compute and I/O on high-performance computing systems has continued to widen, it has become increasingly difficult to perform post-hoc data analytics on full-resolution scientific simulation data due to the high I/O cost. Error-bounded data decomposition and progressive data retrieval framework has recently been developed to address such a challenge by performing data decomposition before storage and reading only part of the decomposed data when necessary. However, the performance of the progressive retrieval framework has been suffering from the over-pessimistic error control theory, such that the achieved maximum error of recomposed data is significantly lower than the required error. Therefore, more data than required is fetched for recomposition, incurring additional I/O overhead. In order to tackle this issue, we propose a DNN-based progressive retrieval framework that can better identify the minimum amount of data to be retrieved. Our contributions are as follows: 1) We provide an in-depth investigation of the recently developed progressive retrieval framework; 2) We propose two designs of prediction models (named D-MGARD and E-MGARD) to estimate the amount of retrieved data size based on error bounds. 3) We evaluate our proposed solutions using scientific datasets generated by real-world simulations from two domains. Evaluation results demonstrate the effectiveness of our solution in accurately predicting the amount of retrieval data size, as well as the advantages of our solution over the traditional approach to reducing the I/O overhead. Based on our evaluation, our solution is shown to read significantly less data (5% - 40% with D-MGARD, 20% - 80% with E-MGARD).

Wang, Jinzhen↗