Search NASA⌕ Search

SEARCH · Search NASA

Results for “physical workload simulation”

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 55 records · Page 3

Modeling Distributed Computing Infrastructures for HEP Applications

Predicting the performance of various infrastructure design options in complex federated infrastructures with computing sites distributed over a wide area network that support a plethora of users and workflows, such as the Worldwide LHC Computing Grid (WLCG), is not trivial. Due to the complexity and size of these infrastructures, it is not feasible to deploy experimental test-beds at large scales merely for the purpose of comparing and evaluating alternate designs. An alternative is to study the behaviours of these systems using simulation. This approach has been used successfully in the past to identify efficient and practical infrastructure designs for High Energy Physics (HEP). A prominent example is the Monarc simulation framework, which was used to study the initial structure of the WLCG. New simulation capabilities are needed to simulate large-scale heterogeneous computing systems with complex networks, data access and caching patterns. A modern tool to simulate HEP workloads that execute on distributed computing infrastructures based on the SimGrid and WRENCH simulation frameworks is outlined. Studies of its accuracy and scalability are presented using HEP as a case-study. Hypothetical adjustments to prevailing computing architectures in HEP are studied providing insights into the dynamics of a part of the WLCG and candidates for improvements.

Horzela, Maximilian↗

Cyber Physical Security (CPS) Extension to Air Traffic Management (ATM) Testbed

The Air Traffic Management (ATM) Testbed is being developed at NASA to enable benefit, impact, safety and cost assessments for accelerating the deployment of Concept and Technologies (C&T) in the National Airspace System (NAS). Today, C&T introduction into the NAS takes decades. The primary reason for this is an inability to assess the operational impact of the interaction between the proposed C&T and operationally deployed systems (Realistic Technologies) in terms of NAS-wide safety, traffic flow efficiency, roles and workload of controllers and traffic managers, and impact on airline fleet operations. Transition of C&T to operations requires mathematical modeling and simulation, Human-in-the-Loop (HITL) testing and shadow-mode evaluation driven by operational data. Whereas interaction with the operational system during testing and stages of deployment is not permissible due to safety concerns, it is certainly possible to create a simulation environment that closely mimics the NAS using the same operational systems/hardware for enabling such assessments. This presentation focuses on a proposed Cyber Physical Security extension to the ATM Testbed for creating a modeling and simulation architecture to study how well the Air Traffic Management system will perform and analyze effectiveness of mitigating security measures against particular cyber-attack scenarios.

Datta, Koushik↗

Strategies for Integrating Deep Learning Surrogate Models with HPC Simulation Applications

The emerging trend of the convergence of high performance computing (HPC), machine learning/deep learning (ML/DL), and big data analytics presents a host of challenges for large-scale computing campaigns that seek best practices to interleave traditional scientific simulation-based workloads with ML/DL models. A portfolio of systematic approaches to incorporate deep learning into modeling and simulation serves a vital need when we support AI for science at a computing facility. In this paper, we evaluate several strategies for deploying deep learning surrogate models in a representative physics application on supercomputers at the Oak Ridge Leadership Computing Facility (OLCF). We discuss a set of recommended deployment architectures and implementation approaches. We analyze and evaluate these alternatives and show their performance and scalability up to 1000 GPUs on two mainstream platforms equipped with different deep learning hardware and software stacks.

Yin, Junqi↗

Evaluating MPI resource usage summary statistics

The Message Passing Interface (MPI) remains the dominant programming model for scientific applications running on today’s high-performance computing (HPC) systems. This dominance stems from MPI’s powerful semantics for inter-process communication that has enabled scientists to write applications for simulating important physical phenomena. MPI does not, however, specify how messages and synchronization should be carried out. Those details are typically dependent on low-level architecture details and the message characteristics of the application. Therefore, analyzing an application’s MPI resource usage is critical to tuning MPI’s performance on a particular platform. The result of this analysis is typically a discussion of the mean message sizes, queue search lengths and message arrival times for a workload or set of workloads. While a discussion of the arithmetic mean in MPI resource usage might be the most intuitive summary statistic, it is not always the most accurate in terms of representing the underlying data. In this paper, we analyze MPI resource usage for a number of key MPI workloads using an existing MPI trace collector and discrete-event simulator. Our analysis demonstrates that the average, while easy and efficient to calculate, is a useful metric for characterizing latency and bandwidth measurements, but may not be a good representation of application message sizes, match list search depths, or MPI inter-operation times. Additionally, we show that the median and mode are superior choices in many cases. We also observe that the arithmetic mean is not the best representation of central tendency for data that are drawn from distributions that are multi-modal or have heavy tails. Furthermore, the results and analysis of our work provide valuable guidance on how we, as a community, should discuss and analyze MPI resource usage data for scientific applications.

97 MATHEMATICS AND COMPUTING↗

Proxy Applications for Converged Workloads: DMC LDRD Initiative

Modern scientific applications are complicated and require coordination of several components. Proxy application driven software-hardware co-design plays a vital role in driving innovation among the developments of applications, software infrastructure and hardware architecture. Proxy applications are self-contained and simplified codes that are intended to model the performance-critical computations within applications. Applications executing on modern High Performance Computing (HPC) systems are susceptible to network congestion, insufficient memory bandwidth within and across compute nodes, and inadvertent loss of performance due to bugs and unoptimized programming models. Modern numerical simulations and machine learning models play a critical role in studying physical phenomenon under myriad uncertainties. Such applications often exhibit irregular computation and memory accesses at specific regions of the application code, which can contribute to various performance bottlenecks at scale. To mitigate such issues and prepare the next generation hardware for a variety of computation and data movement contingencies, a well-known practice is to consider "proxy" applications as representative motifs for various classes of scientific applications. While there is disagreement in the HPC community on the mechanisms of construction of the proxy applications, there is a strong consensus on their positive impact in co-design. Proxy Applications for Converged Workloads (PACER) is about facilitating software-hardware co-design through proxy applications with the goal of improving the performance of converged science workflows on heterogeneous systems.

97 MATHEMATICS AND COMPUTING↗

Constructing a Testing Application for GWMS

Many of Fermilab’s High Energy Physics experiments require High Throughput Computing to carry out simulations, data reconstruction, and data analysis. Glidein Workflow Management System (GWMS) is a tool that distributes High Throughput Computing. The purpose of GWMS is to provide convenient access to Grid resources and sites. GWMS is coupled to HTCondor. HTCondor is the workload management system that is used for scheduling and job control. Users submit jobs to a local HTCondor queue and GWMS will make sure the job will run on one of the many remote resources.

Neely-Brown, LeRayah↗

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↗

The impact of physical and mental tasks on pilot mental workoad

Seven instrument-rated pilots with a wide range of backgrounds and experience levels flew four different scenarios on a fixed-base simulator. The Baseline scenario was the simplest of the four and had few mental and physical tasks. An activity scenario had many physical but few mental tasks. The Planning scenario had few physical and many mental taks. A Combined scenario had high mental and physical task loads. The magnitude of each pilot's altitude and airspeed deviations was measured, subjective workload ratings were recorded, and the degree of pilot compliance with assigned memory/planning tasks was noted. Mental and physical performance was a strong function of the manual activity level, but not influenced by the mental task load. High manual task loads resulted in a large percentage of mental errors even under low mental task loads. Although all the pilots gave similar subjective ratings when the manual task load was high, subjective ratings showed greater individual differences with high mental task loads. Altitude or airspeed deviations and subjective ratings were most correlated when the total task load was very high. Although airspeed deviations, altitude deviations, and subjective workload ratings were similar for both low experience and high experience pilots, at very high total task loads, mental performance was much lower for the low experience pilots.

Berg, S. L.↗

Simulators for quantum network modeling: A comprehensive review

Quantum network research is exploring new networking protocols, physics-based hardware and novel experiments to demonstrate how quantum distribution will work over large distances. Current work explores much of these concepts in simulations, that are developed to understand how quantum networking will be set up and researchers can experiment virtually. Exposing flaws in network designs, like unsustainable topologies, or develop protocols that efficiently utilize network resources, simulators can also help assess whether workloads are balanced across virtual machines in the network. However, much of these simulation models come without reliable verification methods, for testing performance in real deployments. Here, in this paper, we present a review of, to the best of our knowledge, currently used toolkits for modeling quantum networks. With these toolkits and standardized validation techniques, we can lay down the foundations for more accurate and reliable quantum network simulators.

97 MATHEMATICS AND COMPUTING↗

Performance trade-offs in reconfigurable networks for HPC

Designing efficient interconnects to support high-bandwidth and low-latency communication is critical toward realizing high performance computing (HPC) and data center (DC) systems in the exascale era. At extreme computing scales, providing the requisite bandwidth through overprovisioning becomes impractical. These challenges have motivated studies exploring reconfigurable network architectures that can adapt to traffic patterns at runtime using optical circuit switching. Despite the plethora of proposed architectures, surprisingly little is known about the relative performances and trade-offs among different reconfigurable network designs. We aim to bridge this gap by tackling two key issues in reconfigurable network design. First, we study how cost, power consumption, network performance, and scalability vary based on optical circuit switch (OCS) placement in the physical topology. Specifically, we consider two classes of reconfigurable architectures: one that places OCSs between top-of-rack (ToR) switches—ToR-reconfigurable networks (TRNs)—and one that places OCSs between pods of racks—pod-reconfigurable networks (PRNs). Second, we tackle the effects of reconfiguration frequency on network performance. Our results, based on network simulations driven by real HPC and DC workloads, show that while TRNs are optimized for low fan-out communication patterns, they are less suited for carrying high fan-out workloads. PRNs exhibit better overall trade-off, capable of performing comparably to a fully non-blocking fat tree for low fan-out workloads, and significantly outperform TRNs for high fan-out communication patterns.

Teh, Min Yee↗

Telerobotic control of a dextrous manipulator using master and six-DOF hand-controllers for space assembly and servicing tasks

Two studies were conducted evaluating methods of controlling a telerobot; bilateral force reflecting master controllers and proportional rate six degrees of freedom (DOF) hand controllers. The first study compared the controllers on performance of single manipulator arm tasks, a peg-in-the-hole task, and simulated satellite orbital replacement unit changeout. The second study, a Space Station truss assembly task, required simultaneous operation of both manipulator arms (all 12 DOFs) and complex multiaxis slave arm movements. Task times were significantly longer and fewer errors were committed with the hand controllers. The hand controllers were also rated significantly higher in cognitive and manual control workload on the two-arm task. The master controllers were rated significantly higher in physical workload. There were no significant differences in ratings of manipulator control quality.

O'Hara, John M.↗

Evaluation of Aerobic Standards for Lunar Surface Extravehicular Activities

Introduction: As NASA prepares to return to the Moon, astronauts will need to be physically primed to successfully execute Extravehicular Activities (EVA) on the Lunar surface. Compared to past Apollo missions, Artemis missions will include EVAs of increased physical demand, frequency, intensity, and duration, thus requiring adequate fitness to successfully and safely complete mission objectives. The physical demand associated with partial gravity (g) EVAs on the Moon is expected to be greater compared to microgravity EVAs based on initial workload estimation. Currently, aerobic fitness standards for partial g EVAs are not well supported by high-fidelity data and require further research for establishing standards to protect crew health and performance during Lunar surface missions. Therefore, the aim of this investigation is to characterize metabolic data from Lunar analog simulations and in-flight crew population aerobic capacity data to validate the current NASA 3001 standard for celestial partial g aerobic fitness (aerobic capacity (VO_2pk) ≥36.5ml/kg/min). Methods: In order to evaluate aerobic fitness requirements for Lunar EVAs, the following were performed: 1) preliminary analysis of long-duration (6 hr) EVA analog simulations in the Neutral Buoyancy Laboratory (NBL) and the Active Response Gravity Offload System (ARGOS) to evaluate expected metabolic rates for 1/6 g EVAs (NBL: n=1 female; ARGOS: n=1 male) and 2) assessment of the current NASA 3001 celestial surface EVA aerobic standard (aerobic capacity (VO_2pk) ≥36.5ml/kg/min) with data from an ISS astronaut population (n=30 male + 13 female) captured before and during space flight (flight day 15). Preliminary Results: Average fractional aerobic capacity during simulated EVAs were 33%±7% VO_2pk and 23.3±7% VO_2pk in the NBL and ARGOS, respectively. This was within a previously predicted 30–40% sustainable work rate. Average metabolic rates for some tasks performed in the NBL, such as traverse (40.4% VO_2pk) and ingress (47.1% VO_2pk) were higher than the predicted sustainable work range. In ARGOS, the tasks with the greatest metabolic rates were object relocation (34.2% VO_2pk) and incapacitated crew rescue (27.0% VO_2pk). Characterization of ISS crewmember aerobic capacity determined that the average preflight VO_2pk was 42.1±5.4 ml/kg/min for females and 37.5±5.4 ml/kg/min for males. At preflight, 21.5% of crewmembers were below the 36.5 ml/kg/min in-mission aerobic standard for celestial surface EVA as outlined in NASA-STD-3001. In-flight, both female and male crewmembers experienced reductions in VO_2pk (11.7% and 10.9%, respectively), such that, during the mission, 62% of crewmembers were below the standard aerobic capacity level for celestial surface EVAs. Conclusions: Our preliminary data suggest that while average metabolic rates for simulated Lunar EVA fall within the 30–40% sustainable work range, task specific metabolic rates exceed this range and may indicate that greater fitness is necessary for more strenuous tasks expected to be performed on the Lunar surface. Additionally, deconditioning due to space flight results in most crewmembers falling below the current celestial partial g EVA standard, which may increase risk to crew health and performance and completing mission objectives for surface missions. Further research is necessary in Artemis-specific analog environments to validate the current NASA-3001 aerobic standard for celestial EVAs. Additionally, work is ongoing to validate the current NASA-3001 strength standard for celestial EVAs.

N.C. Strock↗

Evaluation of Aerobic Standards for Lunar Surface Extravehicular Activities

As NASA prepares to return to the Moon, astronauts will need to be physically primed to successfully execute Extravehicular Activities (EVA) on the Lunar surface. Compared to past Apollo missions, Artemis missions will include EVAs of increased physical demand, frequency, intensity, and duration, thus requiring adequate fitness to successfully and safely complete mission objectives. The physical demand associated with partial gravity (g) EVAs on the Moon is expected to be greater compared to microgravity EVAs based on initial workload estimation. Currently, aerobic fitness standards for partial g EVAs are not well supported by high-fidelity data and require further research for establishing standards to protect crew health and performance during Lunar surface missions. Therefore, the aim of this investigation is to characterize metabolic data from Lunar analog simulations and in-flight crew population aerobic capacity data to validate the current NASA 3001 standard for celestial partial g aerobic fitness (aerobic capacity (VO_2pk) ≥36.5ml/kg/min).

Nicole Strock↗

The development of a whole-body algorithm

The whole-body algorithm is envisioned as a mathematical model that utilizes human physiology to simulate the behavior of vital body systems. The objective of this model is to determine the response of selected body parameters within these systems to various input perturbations, or stresses. Perturbations of interest are exercise, chemical unbalances, gravitational changes and other abnormal environmental conditions. This model provides for a study of man's physiological response in various space applications, underwater applications, normal and abnormal workloads and environments, and the functioning of the system with physical impairments or decay of functioning components. Many methods or approaches to the development of a whole-body algorithm are considered. Of foremost concern is the determination of the subsystems to be included, the detail of the subsystems and the interaction between the subsystems.

Kay, F. J.↗

Pilot evaluation of experimental flight trajectories in the near-terminal area

Advanced cockpit avionics systems now under development at NASA Langley Research Center will provide the means for effectively utilizing MLS technology to achieve a number of important objectives in the near-terminal airspace. A simulator study has recently been completed at NASA Langley Research Center in which guest pilots from a number of airlines were asked to fly curved ground tracks designed to avoid population centers in four airport communities. Eight two-man crews comprised of commercial line pilots, and three NASA crews each evaluated a total of eight departures and eight approaches on the basis of such factors as workload, safety, passenger acceptance, controllability, and piloting skill. Various physical measurements were also made, including cross-track errors, altitude errors, bank angles, and fuel flow rates. Half of the trajectories were curved and designed to be population-minimal, and half were designed to represent conventional ground tracks. The objective of the study was to determine if statistically significant differences could be detected between curved and conventional ground tracks on the basis of these parameters.

Deloach, R.↗

Initial Investigation of Reaction Control System Design on Spacecraft Handling Qualities for Earth Orbit Docking

A program of research, development, test, and evaluation is planned for the development of Spacecraft Handling Qualities guidelines. In this first experiment, the effects of Reaction Control System design characteristics and rotational control laws were evaluated during simulated proximity operations and docking. Also, the influence of piloting demands resulting from varying closure rates was assessed. The pilot-in-the-loop simulation results showed that significantly different spacecraft handling qualities result from the design of the Reaction Control System. In particular, cross-coupling between translational and rotational motions significantly affected handling qualities as reflected by Cooper-Harper pilot ratings and pilot workload, as reflected by Task-Load Index ratings. This influence is masked but only slightly by the rotational control system mode. While rotational control augmentation using Rate Command Attitude Hold can reduce the workload (principally, physical workload) created by cross-coupling, the handling qualities are not significantly improved. The attitude and rate deadbands of the RCAH introduced significant mental workload and control compensation to evaluate when deadband firings would occur, assess their impact on docking performance, and apply control inputs to mitigate that impact.

Bailey, Randall E.↗

Addressing GPU memory limitations for Graph Neural Networks in High-Energy Physics applications

Introduction Reconstructing low-level particle tracks in neutrino physics can address some of the most fundamental questions about the universe. However, processing petabytes of raw data using deep learning techniques poses a challenging problem in the field of High Energy Physics (HEP). In the Exa.TrkX Project, an illustrative HEP application, preprocessed simulation data is fed into a state-of-art Graph Neural Network (GNN) model, accelerated by GPUs. However, limited GPU memory often leads to Out-of-Memory (OOM) exceptions during training, due to the large size of models and datasets. This problem is exacerbated when deploying models on High-Performance Computing (HPC) systems designed for large-scale applications. Methods We observe a high workload imbalance issue during GNN model training caused by the irregular sizes of input graph samples in HEP datasets, contributing to OOM exceptions. We aim to scale GNNs on HPC systems, by prioritizing workload balance in graph inputs while maintaining model accuracy. Our paper introduces diverse balancing strategies aimed at decreasing the maximum GPU memory footprint and avoiding the OOM exception, across various datasets. Results Our experiments showcase memory reduction of up to 32.14% compared to the baseline. We also demonstrate the proposed strategies can avoid OOM in application. Additionally, we create a distributed multi-GPU implementation using these samplers to demonstrate the scalability of these techniques on the HEP dataset. Discussion By assessing the performance of these strategies as data loading samplers across multiple datasets, we can gauge their effectiveness in both single-GPU and distributed environments. Our experiments, conducted on datasets of varying sizes and across multiple GPUs, broaden the applicability of our work to various GNN applications that handle input datasets with irregular graph sizes.

Lee, Claire Songhyun↗

Fault and Error Latency Under Real Workload: an Experimental Study

A practical methodology for the study of fault and error latency is demonstrated under a real workload. This is the first study that measures and quantifies the latency under real workload and fills a major gap in the current understanding of workload-failure relationships. The methodology is based on low level data gathered on a VAX 11/780 during the normal workload conditions of the installation. Fault occurrence is simulated on the data, and the error generation and discovery process is reconstructed to determine latency. The analysis proceeds to combine the low level activity data with high level machine performance data to yield a better understanding of the phenomena. A strong relationship exists between latency and workload and that relationship is quantified. The sampling and reconstruction techniques used are also validated. Error latency in the memory where the operating system resides was studied using data on the physical memory access. Fault latency in the paged section of memory was determined using data from physical memory scans. Error latency in the microcontrol store was studied using data on the microcode access and usage.

Chillarege, Ram↗