Search NASA⌕ Search

SEARCH · Search NASA

Results for “performance modeling”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2019

PPT-GPU: Scalable GPU Performance Modeling

Performance modeling is a challenging problem due to the complexities of hardware architectures. In this paper, we present PPT-GPU, a scalable and accurate simulation framework that enables GPU code developers and architects to predict the performance of applications in a fast, and accurate manner on different GPU architectures. PPT-GPU is part of the open source project, Performance Prediction Toolkit (PPT) developed at the Los Alamos National Laboratory. We extend the old GPU model in PPT that predict the runtimes of computational physics codes to offer better prediction accuracy, for which, we add models for different memory hierarchies found in GPUs and latencies for different instructions. To further show the utility of PPT-GPU, we compare our model against real GPU device(s) and the widely used cycle-accurate simulator, GPGPU-Sim using different workloads from RODINIA and Parboil benchmarks. The results indicate that the predicted performance of PPT-GPU is within a 10% error compared to the real device(s). In addition, PPT-GPU is highly scalable, where it is up to 450x faster than GPGPU-Sim with more accurate results.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1989

Iterative procedures for space shuttle main engine performance models

Performance models of the Space Shuttle Main Engine (SSME) contain iterative strategies for determining approximate solutions to nonlinear equations reflecting fundamental mass, energy, and pressure balances within engine flow systems. Both univariate and multivariate Newton-Raphson algorithms are employed in the current version of the engine Test Information Program (TIP). Computational efficiency and reliability of these procedures is examined. A modified trust region form of the multivariate Newton-Raphson method is implemented and shown to be superior for off nominal engine performance predictions. A heuristic form of Broyden's Rank One method is also tested and favorable results based on this algorithm are presented.

Santi, L. Michael↗
NASA NTRS2017

Mars 2020 Surface Mission Performance Modeling: Part 3. Mission Performance Modeling Approach and Results

We have developed a comprehensive model of the Mars 2020 rover surface mission used to identify and evaluate mission productivity and operability challenges and opportunities. The surface mission performance model combines many aspects of rover design, operations approach, system behaviors, and mission constraints into an integrated model of surface mission performance. The surface mission model applies monte-carlo methodologies in order to investigate the key sources of mission performance variability and their effects with respect to overall mission performance. Our modeling efforts aims to investigate and inform engineering design and development efforts, as well as the landing site selection process led by the Mars science community. We believe that an integrated mission performance model such as this is important to aligning engineering and science efforts across the project.

Ono, Masahiro↗
DOE OSTI2016

Performance Modeling of In Situ Rendering

With the push to exascale, in situ visualization and analysis will continue to play an important role in high performance computing. Tightly coupling in situ visualization with simulations constrains resources for both, and these constraints force a complex balance of trade-offs. A performance model that provides an a priori answer for the cost of using an in situ approach for a given task would assist in managing the trade-offs between simulation and visualization resources. In this work, we present new statistical performance models, based on algorithmic complexity, that accurately predict the run-time cost of a set of representative rendering algorithms, an essential in situ visualization task. To train and validate the models, we conduct a performance study of an MPI+X rendering infrastructure used in situ with three HPC simulation applications. We then explore feasibility issues using the model for selected in situ rendering questions.

Larsen, Matthew↗
DOE OSTI2016

Development and validation of chemistry agnostic flow battery cost performance model and application to nonaqueous electrolyte systems: Chemistry agnostic flow battery cost performance model

A chemistry agnostic cost performance model is described for a nonaqueous flow battery. The model predicts flow battery performance by estimating the active reaction zone thickness at each electrode as a function of current density, state of charge, and flow rate using measured data for electrode kinetics, electrolyte conductivity, and electrode-specific surface area. Validation of the model is conducted using a 4kW stack data at various current densities and flow rates. This model is used to estimate the performance of a nonaqueous flow battery with electrode and electrolyte properties used from the literature. The optimized cost for this system is estimated for various power and energy levels using component costs provided by vendors. The model allows optimization of design parameters such as electrode thickness, area, flow path design, and operating parameters such as power density, flow rate, and operating SOC range for various application duty cycles. A parametric analysis is done to identify components and electrode/electrolyte properties with the highest impact on system cost for various application durations. A pathway to 100$kWh -1 for the storage system is identified.

25 ENERGY STORAGE↗
DOE OSTI2017

PV Performance Modeling Methods and Practices: Results from the 4th PV Performance Modeling Collaborative Workshop.

In 2014, the IEA PVPS Task 13 added the PVPMC as a formal activity to its technical work plan for 2014-2017. The goal of this activity is to expand the reach of the PVPMC to a broader international audience and help to reduce PV performance modeling uncertainties worldwide. One of the main deliverables of this activity is to host one or more PVPMC workshops outside the US to foster more international participation within this collaborative group. This report reviews the results of the first in a series of these joint IEA PVPS Task 13/PVPMC workshops. The 4th PV Performance Modeling Collaborative Workshop was held in Cologne, Germany at the headquarters of TÜV Rheinland on October 22-23, 2015.

14 SOLAR ENERGY↗
NASA NTRS2011

Predicting the Consequences of Workload Management Strategies with Human Performance Modeling

Human performance modelers at the US Army Research Laboratory have developed an approach for establishing Soldier high workload that can be used for analyses of proposed system designs. Their technique includes three key components. To implement the approach in an experiment, the researcher would create two experimental conditions: a baseline and a design alternative. Next they would identify a scenario in which the test participants perform all their representative concurrent interactions with the system. This scenario should include any events that would trigger a different set of goals for the human operators. They would collect workload values during both the control and alternative design condition to see if the alternative increased workload and decreased performance. They have successfully implemented this approach for military vehicle. designs using the human performance modeling tool, IMPRINT. Although ARL researches use IMPRINT to implement their approach, it can be applied to any workload analysis. Researchers using other modeling and simulations tools or conducting experiments or field tests can use the same approach.

Mitchell, Diane Kuhl↗
DOE OSTI2016

A Performance Model for Allocating the Parallelism in a Multigrid-in-Time Solver

The traditional way to numerically solve time-dependent problems is to sequentially march through time, solving for one time step and then the next. The parallelism in this approach is limited to the spatial dimension, which is quickly exhausted, causing the gain from using more processors to solve a problem to diminish. One approach to overcome this barrier is to use methods that are parallel in time. These methods have the potential to achieve dramatically better performance compared to time-stepping approaches, but achieving this performance requires carefully choosing the amount of parallelism devoted to space versus the amount devoted to time. Here, we present a performance model that, for a multigrid-in-time solver, makes the decision on when to switch to parallel-in-time and on how much parallelism to devote to space vs. time. In our experiments, the model selects the best parallel configuration in most of our test cases and a configuration close to the best one in all other cases.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Adaptive Learning for Concept Drift in Application Performance Modeling

Supervised learning is a promising approach for modeling the performance of applications running on large HPC systems. A key assumption in supervised learning is that the training and testing data are obtained under the same conditions. However, in production HPC systems these conditions might not hold because the conditions of the platform can change over time as a result of hardware degradation, hardware replacement, software upgrade, and configuration updates. These changes could alter the data distribution in a way that affects the accuracy of the predictive performance models and render them less useful; this phenomenon is referred to as concept drift. Ignoring concept drift can lead to suboptimal resource usage and decreased efficiency when those performance models are deployed for tuning and job scheduling in production systems. To address this issue, we propose a concept-drift-aware predictive modeling approach that comprises two components: (1) an online Bayesian changepoint detection method that can automatically identify the location of events that lead to concept drift in near-real time and (2) a moment-matching transformation inspired by transfer learning that converts the training data collected before the drift to be useful for retraining. We use application input/output performance data collected on Cori, a production supercomputing system at the National Energy Research Scientific Computing Center, to demonstrate the effectiveness of our approach. The results show that concept-drift-aware models obtain significant improvement in accuracy; the median absolute error of the best-performing Gaussian process regression improved by 58.8% when the proposed approaches were used.

adaptive learning↗
DOE OSTI2024

Fuel Performance Modeling Status Update and Potential Model Improvements

Fuel performance modeling status update and potential model improvements overview of TRISO fuel performance modeling codes PARFUME/BISON, AGR experiment support, potential modeling improvements, BISON smeared cracking model, BISON fission product source term, and AGR-3/4 reirradiation heating test.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
NASA NTRS1989

Predictive performance models and multiple task performance

Five models that predict how performance of multiple tasks will interact in complex task scenarios are discussed. The models are shown in terms of the assumptions they make about human operator divided attention. The different assumptions about attention are then empirically validated in a multitask helicopter flight simulation. It is concluded from this simulation that the most important assumption relates to the coding of demand level of different component tasks.

Wickens, Christopher D.↗
DOE OSTI2021

Design-time performance modeling of compositional parallel programs

Performance models are powerful instruments for understanding the performance of parallel systems and uncovering their bottlenecks. Already during system design, performance models can help ponder alternative development options. However, creating a performance model – whether theoretically or empirically – for an entire application that does not exist yet is challenging. In this paper, we propose to generate performance models of full programs from performance models of their components using formal composition operators derived from parallel design patterns. As long as the design of the overall system follows such a pattern, its performance model can be predicted with reasonable accuracy without an actual implementation. In conclusion, we demonstrate our approach with design patterns of varying complexity, including pipeline, task pool, and eventually MapReduce, which is representative of a broad class of data-analytics applications.

97 MATHEMATICS AND COMPUTING↗