Search NASA⌕ Search

SEARCH · Search NASA

Results for “timing error evaluation”

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

Statistically Appraising Process Quality of Affinity Isolation Experiments

Quality affinity isolation experiments are necessary to identify valid protein-protein interactions. Biological error, processing error, and random variability can reduce the quality of an experiment, and thus hinder the identification of protein interaction pairs. Appraising affinity isolation assay quality is essential to inferring protein associations. An important step of the assay is the mass spectrometric identification of proteins. To evaluate this step, a known mixture of proteins is processed through a mass spectrometer as a quality control mixture. If the mass spectrometer yields unexpected results, the process is currently qualitatively evaluated, tuned, and reset. Statistical quality control (SQC) procedures, including the use of cumulative sum, the individual measurement, and moving range charts are implemented to analyze the stability of the mass spectrometric analysis. The SQC measures presented here can assist in establishing preliminary control limits to identify an out-of-control process and investigate assignable causes for shifts in the process mean in real-time.

59 BASIC BIOLOGICAL SCIENCES↗

Automated Voxel-Based Analysis of Volumetric Dynamic Contrast-Enhanced CT Data Improves Measurement of Serial Changes in Tumor Vascular Biomarkers

Objectives: Development of perfusion imaging as a biomarker requires more robust methodologies for quantification of tumor physiology that allow assessment of volumetric tumor heterogeneity over time. This study proposes a parametric method for automatically analyzing perfused tissue from volumetric dynamic contrast-enhanced (DCE) computed tomography (CT) scans and assesses whether this 4-dimensional (4D) DCE approach is more robust and accurate than conventional, region-of-interest (ROI)-based CT methods in quantifying tumor perfusion with preliminary evaluation in metastatic brain cancer. Methods and Materials: Functional parameter reproducibility and analysis of sensitivity to imaging resolution and arterial input function were evaluated in image sets acquired from a 320-slice CT with a controlled flow phantom and patients with brain metastases, whose treatments were planned for stereotactic radiation surgery and who consented to a research ethics board-approved prospective imaging biomarker study. A voxel-based temporal dynamic analysis (TDA) methodology was used at baseline, at day 7, and at day 20 after treatment. The ability to detect changes in kinetic parameter maps in clinical data sets was investigated for both 4D TDA and conventional 2D ROI-based analysis methods. Results: A total of 7 brain metastases in 3 patients were evaluated over the 3 time points. The 4D TDA method showed improved spatial efficacy and accuracy of perfusion parameters compared to ROI-based DCE analysis (P<.005), with a reproducibility error of less than 2% when tested with DCE phantom data. Clinically, changes in transfer constant from the blood plasma into the extracellular extravascular space (K{sub trans}) were seen when using TDA, with substantially smaller errors than the 2D method on both day 7 post radiation surgery (±13%; P<.05) and by day 20 (±12%; P<.04). Standard methods showed a decrease in K{sub trans} but with large uncertainty (111.6 ± 150.5) %. Conclusions: Parametric voxel-based analysis of 4D DCE CT data resulted in greater accuracy and reliability in measuring changes in perfusion CT-based kinetic metrics, which have the potential to be used as biomarkers in patients with metastatic brain cancer.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Dynamic estimation of three-dimensional cerebrovascular deformation from rotational angiography

Purpose: The objective of this study is to investigate the feasibility of detecting and quantifying 3D cerebrovascular wall motion from a single 3D rotational x-ray angiography (3DRA) acquisition within a clinically acceptable time and computing from the estimated motion field for the further biomechanical modeling of the cerebrovascular wall. Methods: The whole motion cycle of the cerebral vasculature is modeled using a 4D B-spline transformation, which is estimated from a 4D to 2D+t image registration framework. The registration is performed by optimizing a single similarity metric between the entire 2D+t measured projection sequence and the corresponding forward projections of the deformed volume at their exact time instants. The joint use of two acceleration strategies, together with their implementation on graphics processing units, is also proposed so as to reach computation times close to clinical requirements. For further characterizing vessel wall properties, an approximation of the wall thickness changes is obtained through a strain calculation. Results: Evaluation on in silico and in vitro pulsating phantom aneurysms demonstrated an accurate estimation of wall motion curves. In general, the error was below 10% of the maximum pulsation, even in the situation when substantial inhomogeneous intensity pattern was present. Experiments on in vivo data provided realistic aneurysm and vessel wall motion estimates, whereas in regions where motion was neither visible nor anatomically possible, no motion was detected. The use of the acceleration strategies enabled completing the estimation process for one entire cycle in 5-10 min without degrading the overall performance. The strain map extracted from our motion estimation provided a realistic deformation measure of the vessel wall. Conclusions: The authors' technique has demonstrated that it can provide accurate and robust 4D estimates of cerebrovascular wall motion within a clinically acceptable time, although it has to be applied to a larger patient population prior to possible wide application to routine endovascular procedures. In particular, for the first time, this feasibility study has shown that in vivo cerebrovascular motion can be obtained intraprocedurally from a 3DRA acquisition. Results have also shown the potential of performing strain analysis using this imaging modality, thus making possible for the future modeling of biomechanical properties of the vascular wall.

60 APPLIED LIFE SCIENCES↗

Leveraging Machine Learning Capabilities for the Characterization of Irradiated Uranium: A Case Study of Analysis Methods for Nuclear Safeguards and Nuclear Forensics

Nondestructively determining the initial enrichment of irradiated uranium is a complex and laborious multivariable problem due to the presence of fission products. This work demonstrates the capabilities of machine learning to analyze gamma-ray spectral data to determine initial enrichment without knowledge of the decay time of the sample. The approach developed is agnostic to the particular scenario and is applicable to a wide variety of applications in nuclear forensics and nuclear safeguards. We irradiated 5 mg uranium standard reference materials at discrete enrichment values ranging from 0.02% to 97% 235 U (weight percent) in UT Austin’s Nuclear Engineering Teaching Laboratory TRIGA Mark II 1.1 MW research reactor, allowed each to decay for 8 hours, and then measured each sample via gamma-ray spectrometry for 50 hours post-irradiation yielding 1,400 individual gamma-ray spectra discretized into 8,192 energy bins. We then trained decision trees models to analyze individual gamma-ray spectra and estimate the associated initial enrichment without knowledge of the time since end of irradiation. We evaluated the performance of the models with a reserved test set not used for training or calibrating the model. A decision tree model constructed with this procedure achieved a mean absolute error in initial enrichment determination of 2.3% (weight percent 235 U). Next, we implemented a principal component analysis pre-processing routine of the gamma-ray spectrometry data to reduce the dimensionality of the dataset from 8,192 channels in the spectrum to 10 principal components while retaining over 99% of the inherent variance in the data. Decision tree models constructed with these data demonstrated decreased mean absolute error in enrichment determination, reduced computational time, and decreased complexity. A single decision tree model constructed with this procedure achieved a mean absolute error in initial enrichment determination of 0.05% (weight percent 235 U). Furthermore, we analyzed these models with learning curves to ensure that overfitting did not occur. The capabilities provided by these models can be naturally extended to other application-focused measurements in the fields of nuclear safeguards, nuclear forensics, and nuclear non-proliferation.

Drescher, Adam↗

Effects of error covariance structure on estimation of model averaging weights and predictive performance

When conducting model averaging for assessing groundwater conceptual model uncertainty, the averaging weights are often evaluated using model selection criteria such as AIC, AICc, BIC, and KIC (Akaike Information Criterion, Corrected Akaike Information Criterion, Bayesian Information Criterion, and Kashyap Information Criterion, respectively). However, this method often leads to an unrealistic situation in which the best model receives overwhelmingly large averaging weight (close to 100%), which cannot be justified by available data and knowledge. It was found in this study that this problem was caused by using the covariance matrix, CE, of measurement errors for estimating the negative log likelihood function common to all the model selection criteria. This problem can be resolved by using the covariance matrix, Cek, of total errors (including model errors and measurement errors) to account for the correlation between the total errors. An iterative two-stage method was developed in the context of maximum likelihood inverse modeling to iteratively infer the unknown Cek from the residuals during model calibration. The inferred Cek was then used in the evaluation of model selection criteria and model averaging weights. While this method was limited to serial data using time series techniques in this study, it can be extended to spatial data using geostatistical techniques. The method was first evaluated in a synthetic study and then applied to an experimental study, in which alternative surface complexation models were developed to simulate column experiments of uranium reactive transport. It was found that the total errors of the alternative models were temporally correlated due to the model errors. The iterative two-stage method using Cek resolved the problem that the best model receives 100% model averaging weight, and the resulting model averaging weights were supported by the calibration results and physical understanding of the alternative models. Using Cek obtained from the iterative two-stage method also improved predictive performance of the individual models and model averaging in both synthetic and experimental studies.

error correlation↗

Design of set-point weighting PI{sup λ} + D{sup μ} controller for vertical magnetic flux controller in Damavand tokamak

In this paper, a simple method is presented for tuning weighted PI{sup λ} + D{sup μ} controller parameters based on the pole placement controller of pseudo-second-order fractional systems. One of the advantages of this controller is capability of reducing the disturbance effects and improving response to input, simultaneously. In the following sections, the performance of this controller is evaluated experimentally to control the vertical magnetic flux in Damavand tokamak. For this work, at first a fractional order model is identified using output-error technique in time domain. For various practical experiments, having desired time responses for magnetic flux in Damavand tokamak, is vital. To approach this, at first the desired closed loop reference models are obtained based on generalized characteristic ratio assignment method in fractional order systems. After that, for the identified model, a set-point weighting PI{sup λ} + D{sup μ} controller is designed and simulated. Finally, this controller is implemented on digital signal processor control system of the plant to fast/slow control of magnetic flux. The practical results show appropriate performance of this controller.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Evaluation of a pair-based, joint-likelihood association approach for regional infrasound event identification

SUMMARY A Bayesian framework for the association of infrasonic detections is presented and evaluated for analysis at regional propagation scales. A pair-based, joint-likelihood association approach is developed that identifies events by computing the probability that individual detection pairs are attributable to a hypothetical common source and applying hierarchical clustering to identify events from the pair-based analysis. The framework is based on a Bayesian formulation introduced for infrasonic source localization and utilizes the propagation models developed for that application with modifications to improve the numerical efficiency of the analysis. Clustering analysis is completed using hierarchical analysis via weighted linkage for a non-Euclidean distance matrix defined by the negative log-joint-likelihood values. The method is evaluated using regional synthetic data with propagation distances of hundreds of kilometres in order to study the sensitivity of the method to uncertainties and errors in backazimuth and time of arrival. The method is found to be robust and stable for typical uncertainties, able to effectively distinguish noise detections within the data set from those in events, and can be made numerically efficient due to its ease of parallelization.

58 GEOSCIENCES↗

Building a Wide-Area File Transfer Performance Predictor: An Empirical Study

Wide-area data transfer is central to geographically distributed scientific workflows. Faster delivery of data is important for these workflows. Predictability is equally (or even more) important. With the goal of providing a reasonably accurate estimate of data transfer time to improve resource allocation & scheduling for workflows and enable end-to-end data transfer optimization, we apply machine learning methods to develop predictive models for data transfer times over a variety of wide area networks. To build and evaluate these models, we use 201,388 transfers, involving 759 million files totaling 9 PB transferred, over 115 heavily used source-destination pairs (“edges”) between 135 unique endpoints. We evaluate the models for different retraining frequencies and different window size of history data. In the best case, the resulting models have a median prediction error of ≤21% for 50% of the edges, and ≤32% for 75% of the edges. We present a detailed analysis of these results that provides insights into the cause of some of the high errors. We envision that the performance predictor will be informative for scheduling geo-distributed workflows. The insights also suggest obvious directions for both further analysis and transfer service optimization.

Liu, Zhengchun↗

An Approach to Dependence Assessment in Human Reliability Analysis: Application of Lag and Linger Effects

Dependence assessment refers to an approach used in human reliability analysis (HRA) to adjust a human error probability (HEP) for the following action by considering the impact of the preceding action. It has been known to significantly affect the overall results of probabilistic safety assessment (PSA). If the dependence assessment is not adequate, the result could be unconvincing for explaining the operator failures in the context of PSA. To date, several methods and some recent research have identified suggestions for treating dependence issues in HRA; however, these are still exclusively based on the intrinsic approach of the Technique for Human Error Rate Prediction (THERP), an HRA method. THERP inevitably has a challenge with the subjectivity of expert evaluation as well as the requirement for PSA and HRA expertise with resource-intensive and time-consuming processes. This paper suggests an approach to dependence assessment that could not only minimize the influence of expert judgment, but also saves time to perform the analysis with reasonable manpower. It modifies existing HRA methods with considering lag and linger effects to apply dependence effects for them. Based on a representative HRA method, i.e., Standardized Plant Analysis Risk - HRA (SPAR-H), guidance for how to apply lag and linger effects for the HRA method is suggested. Then, an investigation is carried out to compare quantification results of the revised HRA method with that of the original approach based on experimental data.

99 GENERAL AND MISCELLANEOUS↗

Cumulative dose on fractional delivery of tomotherapy to periodically moving organ: A phantom QA suggestion

This study was conducted to evaluate the cumulative dosimetric error that occurs in both target and surrounding normal tissues when treating a moving target in multifractional treatment with tomotherapy. An experiment was devised to measure cumulative error in multifractional treatments delivered to a horseshoe-shaped clinical target volume (CTV) surrounding a cylinder shape of organ at risk (OAR). Treatments differed in jaw size (1.05 vs 2.5 cm), pitch (0.287 vs 0.660), and modulation factor (1.5 vs 2.5), and tumor motion characteristics differing in amplitude (1 to 3 cm), period (3 to 5 second), and regularity (sinusoidal vs irregular) were tested. Treatment plans were delivered to a moving phantom up to 5-times exposure. Dose distribution on central coronal plane from 1 to 5 times exposure was measured with GAFCHROMIC EBT film. Dose differences occurring across 1 to 5 times exposure of treatment and between treatment plans were evaluated by analyzing measurements of gamma index, gamma index histogram, histogram changes, and dose at the center of the OAR. The experiment showed dose distortion due to organ motion increased between multiexposure 1 to 3 times but plateaued and remained constant after 3-times exposure. In addition, although larger motion amplitude and a longer period of motion both increased dosimetric error, the dose at the OAR was more significantly affected by motion amplitude rather than motion period. Irregularity of motion did not contribute significantly to dosimetric error when compared with other motion parameters. Restriction of organ motion to have small amplitude and short motion period together with larger jaw size and small modulation factor (with small pitch) is effective in reducing dosimetric error. Pretreatment measurements for 3-times exposure of treatment to a moving phantom with patient-specific tumor motion would provide a good estimation of the delivered dose distribution.

61 RADIATION PROTECTION AND DOSIMETRY↗

TH-AB-202-08: A Robust Real-Time Surface Reconstruction Method On Point Clouds Captured From a 3D Surface Photogrammetry System

Purpose: Surface photogrammetry (e.g. VisionRT, C-Rad) provides a noninvasive way to obtain high-frequency measurement for patient motion monitoring in radiotherapy. This work aims to develop a real-time surface reconstruction method on the acquired point clouds, whose acquisitions are subject to noise and missing measurements. In contrast to existing surface reconstruction methods that are usually computationally expensive, the proposed method reconstructs continuous surfaces with comparable accuracy in real-time. Methods: The key idea in our method is to solve and propagate a sparse linear relationship from the point cloud (measurement) manifold to the surface (reconstruction) manifold, taking advantage of the similarity in local geometric topology in both manifolds. With consistent point cloud acquisition, we propose a sparse regression (SR) model to directly approximate the target point cloud as a sparse linear combination from the training set, building the point correspondences by the iterative closest point (ICP) method. To accommodate changing noise levels and/or presence of inconsistent occlusions, we further propose a modified sparse regression (MSR) model to account for the large and sparse error built by ICP, with a Laplacian prior. We evaluated our method on both clinical acquired point clouds under consistent conditions and simulated point clouds with inconsistent occlusions. The reconstruction accuracy was evaluated w.r.t. root-mean-squared-error, by comparing the reconstructed surfaces against those from the variational reconstruction method. Results: On clinical point clouds, both the SR and MSR models achieved sub-millimeter accuracy, with mean reconstruction time reduced from 82.23 seconds to 0.52 seconds and 0.94 seconds, respectively. On simulated point cloud with inconsistent occlusions, the MSR model has demonstrated its advantage in achieving consistent performance despite the introduced occlusions. Conclusion: We have developed a real-time and robust surface reconstruction method on point clouds acquired by photogrammetry systems. It serves an important enabling step for real-time motion tracking in radiotherapy. This work is supported in part by NIH grant R01 CA169102-02.

60 APPLIED LIFE SCIENCES↗

TU-EF-304-03: 4D Monte Carlo Robustness Test for Proton Therapy

Purpose: Breathing motion and approximate dose calculation engines may increase proton range uncertainties. We address these two issues using a comprehensive 4D robustness evaluation tool based on an efficient Monte Carlo (MC) engine, which can simulate breathing with no significant increase in computation time. Methods: To assess the robustness of the treatment plan, multiple scenarios of uncertainties are simulated, taking into account the systematic and random setup errors, range uncertainties, and organ motion. Our fast MC dose engine, called MCsquare, implements optimized models on a massively-parallel computation architecture and allows us to accurately simulate a scenario in less than one minute. The deviations of the uncertainty scenarios are then reported on a DVH-band and compared to the nominal plan.The robustness evaluation tool is illustrated in a lung case by comparing three 60Gy treatment plans. First, a plan is optimized on a PTV obtained by extending the CTV with an 8mm margin, in order to take into account systematic geometrical uncertainties, like in our current practice in radiotherapy. No specific strategy is employed to correct for tumor and organ motions. The second plan involves a PTV generated from the ITV, which encompasses the tumor volume in all breathing phases. The last plan results from robust optimization performed on the ITV, with robustness parameters of 3% for tissue density and 8 mm for positioning errors. Results: The robustness test revealed that the first two plans could not properly cover the target in the presence of uncertainties. CTV-coverage (D95) in the three plans ranged respectively between 39.4–55.5Gy, 50.2–57.5Gy, and 55.1–58.6Gy. Conclusion: A realistic robustness verification tool based on a fast MC dose engine has been developed. This test is essential to assess the quality of proton therapy plan and very useful to study various planning strategies for mobile tumors. This work is partly funded by IBA (Louvain-la-Neuve, Belgium)

60 APPLIED LIFE SCIENCES↗

Short-term electricity load forecasting: Application-driven evaluation of machine learning models across spatial and temporal scales

As we transition towards a decarbonized economy, the integration of variable renewable energy resources and new demands (e.g., electric vehicles, heat pumps) into the electricity grid places unprecedented pressure on grid operators to effectively anticipate and manage peak load. In this context, machine learning algorithms are proving to be indispensable for accurate short-term load forecasting, a crucial task to address these challenges. This study benchmarks 6 machine learning algorithms, including three neural networks and three tree-based algorithms, across various levels of spatial aggregation and time horizons (1, 4, 8, 24, and 48 h). The central contribution of this work is the comparison and analysis of load forecasting models not only based on statistical metrics, but also based on a novel error metric, which evaluates the cost implications of forecast errors for power system stakeholders. Results show that tree-based models outperform neural networks, based on statistical metrics, and yield less skewed error distributions for most spatial scales. However, through the lens of the novel error metric, neural networks are the more competitive choice, especially for forecast horizons that exceed 8 h. The study concludes with actionable recommendations to grid operators and highlights the need for the development of error metrics that link forecasting accuracy to operational costs. To promote transparency and open science, the datasets and Python code are open-sourced via a supplementary repository.

Houben, Nikolaus↗

Development of Streamlined Nuclear Safety Analysis Tool for Spent Nuclear Fuel Applications

To understand the changing nuclear and mechanical characteristics of spent nuclear fuel (SNF) or used nuclear fuel (UNF) and the different storage, transportation, and disposal systems at various stages within the waste management system, different types of analyses are required. These analyses require the use of assorted tools and numerous types of data. Using the appropriate modeling and simulation (M&S) parameters and selecting from the diversity of analytic tools to conduct SNF analyses can be a tedious, error-prone, and time-consuming undertaking for analysts and reviewers alike. A new, integrated data and analysis system was designed to simplify and automate performance of accurate, efficient evaluations for characterizing the input to the overall U.S. nuclear waste management system—the UNF-Storage, Transportation & Disposal Analysis Resource and Data System (UNF-ST&DARDS). A relational database has been assembled to provide a standard means by which UNF-ST&DARDS can succinctly store and retrieve M&S parameters for specific SNF analysis. A library of various analysis model templates is used to communicate M&S parameters for the most appropriate M&S application. A process manager facilitates performance of actual as-loaded, assembly-specific, and cask-specific evaluations. Interactive visualization capabilities facilitate data analysis and results interpretation. To date, UNF-ST&DARDS has completed (1) explicit depletion and decay analysis of every fuel assembly (~245 000) discharged from commercial U.S. reactors through June 2013, with 13 cooling time steps (results include isotopic compositions for 142 isotopes, and radiation and thermal source terms); (2) SNF radiation dose rate evaluations at 1 m for all the fuel assemblies discharged through June 2013; and (3) criticality, shielding, thermal, and containment analyses of hundreds of loaded casks. In conclusion, UNF-ST&DARDS also provides various automated report generation capabilities with dynamic figure and table update capabilities based on changes to the Unified Database.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Minicharged particles from the Sun: a cutting-edge bound

We study the impact on the Sun of an exotic energy-loss channel caused by plasmon decay into fermionic minicharged particles with charge ε e and mass m{sub f}. We compare solar models with this extra emission to helioseismological and neutrino data, obtaining a bound ε < 2.2 × 10{sup −14} (95% CL) for m{sub f} ∼< 25 eV. Our result is comparable to previous limits from the cooling of globular cluster stars, while at the same time it is better understood and takes theoretical and observational errors into account.

79 ASTRONOMY AND ASTROPHYSICS↗

Methodology for Wide Band-Gap Device Dynamic Characterization

Here, the double pulse test (DPT) is a widely accepted method to evaluate the dynamic behavior of power devices. Considering the high switching-speed capability of wide band-gap devices, the test results are very sensitive to the alignment of voltage and current (V-I) measurements. Also, because of the shoot-through current induced by Cdv/dt (i.e., cross-talk), the switching losses of the nonoperating switch device in a phase-leg must be considered in addition to the operating device. This paper summarizes the key issues of the DPT, including components and layout design, measurement considerations, grounding effects, and data processing. Additionally, a practical method is proposed for phase-leg switching loss evaluation by calculating the difference between the input energy supplied by a dc capacitor and the output energy stored in a load inductor. Based on a phase-leg power module built with 1200-V/50-A SiC MOSFETs, the test results show that this method can accurately evaluate the switching loss of both the upper and lower switches by detecting only one switching current and voltage, and it is immune to V-I timing misalignment errors.

30 DIRECT ENERGY CONVERSION↗

Profiling and Predicting Application Performance on the Cloud

Cloud providers continue to expand and diversify their collection of leasable resources to meet the needs of an increasingly wide range of applications. While this flexibility is a key benefit of the cloud, it also creates a complex landscape in which users are faced with many resource choices for a given application. Suboptimal selections can both degrade performance and increase costs. Given the rapidly evolving pool of resources, it is infeasible for users alone to select instance types; instead, automated methods are needed to simplify and guide resource provisioning. Here we present a method for the automatic prediction of application performance on arbitrary cloud instances. We combine offline and online profiling approaches, using historical data gathered from non-cloud environments and targeted profiling runs on cloud environments to create a composite application model that can predict run times on a given cloud instance type for a given input data size. We demonstrate average error of 17.2% across nine applications used in production bioinformatics workflows. Finally, we evaluate an experiment design approach to explore the trade-off between the cost of profiling and the accuracy of our models. Using this approach, with no prior knowledge, we show that using 4 selectively chosen experiments we can achieve performance within 30% of a model trained using all instance types.

Cloud profiling↗

Machine Learning Assisted HPC Workload Trace Generation for Leadership Scale Storage Systems

Monitoring and analyzing a wide range of I/O activities in an HPC cluster is important in maintaining mission-critical performance in a large-scale, multi-user, parallel storage system. Center-wide I/O traces can provide high-level information and fine-grained activities per application or per user running in the system. Studying such large-scale traces can provide helpful insights into the system. It can be used to develop predictive methods for making predictive decisions, adjusting scheduling policies, or providing decisions for the design of next-generation systems. However, sharing real-world I/O traces to expedite such research efforts leaves a few concerns; i) the cost of sharing the large traces is expensive due to this large size, and ii) privacy concern is an issue.We address such issues by building an end-to-end machine learn- ing (ML) workflow that can generate I/O traces for large-scale HPC applications. We leverage ML based feature selection and gener- ative models for I/O trace generation. The generative models are trained on I/O traces collected by the darshan I/O characterization tool over a period of one year. We present a two-step generation process consisting of two deep-learning models, called the feature generator and the trace generator. The combination of two-step generative models provides robustness by reducing the bias of the model and accounting for the stochastic nature of the I/O traces across different runs of an application. We evaluate the performance of the generative models and show that the two-step model can generate time-series I/O traces with less than 20% root mean square error.

Paul, Arnab↗