Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “computer system performance”

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 1,117 records · Page 62

Model description document for a computer program for the emulation/simulation of a space station environmental control and life support system (ESCM)

Emulation/Simulation Computer Model (ESCM) computes the transient performance of a Space Station air revitalization subsystem with carbon dioxide removal provided by a solid amine water desorbed subsystem called SAWD. This manual describes the mathematical modeling and equations used in the ESCM. For the system as a whole and for each individual component, the fundamental physical and chemical laws which govern their operations are presented. Assumptions are stated, and when necessary, data is presented to support empirically developed relationships.

Yanosy, James L.↗

High-performance network and channel based storage

In the traditional mainframe-centered view of a computer system, storage devices are coupled to the system through complex hardware subsystems called I/O channels. With the dramatic shift toward workstation-based computing, and its associated client/server model of computation, storage facilities are now found attached to file servers and distributed throughout the network. In this paper, we discuss the underlying technology trends that are leading to high-performance network-based storage, namely advances in networks, storage devices, and I/O controller and server architectures. We review several commercial systems and research prototypes that are leading to a new approach to high-performance computing based on network-attached storage.

Katz, Randy H.↗

Future SDN-HPON Control Plane Architecture and Protocol for On-Demand Terabit End-to-End Extreme-Scale Science Applications

The UC Davis team conducted architecture, algorithm and experimental studies for next generation ultra-high-bandwidth optical networks in support of extreme-scale science applications. In particular, over the course of this 3-year project, the UC Davis team achieved the following main accomplishments: Designed a network architecture for automated multi-domain service provisioning: Designed information exchange schemes between autonomous systems (domains) Designed application programming interfaces to orchestrate resource reservation in data centers, high performance computation facilities, and communication networks belonging to different autonomous systems Experimentally assessed the multi-domain network architecture in a distributed field trial set-up connecting premises in three continents Designed a broker-based orchestrated multi-domain optical networking framework supporting observe-analyze-act cycle based cognitive inter-domain service provisioning: Designed an inter-domain operation mechanism, e.g., topology abstraction, for the broker plane and domain managers to collaboratively accomplish inter-domain service provisioning while respecting the autonomy of each administrative domain Designed an alien wavelength monitoring scheme and an artificial neural network based optical signal-to-noise ratio estimator for quality-of-transmission aware inter-domain service provisioning Developed a deep neural network based multi-domain traffic predictor and a cognitive routing and spectrum assignment algorithm to realize knowledge-based autonomous traffic engineering in multi-domain networks Implemented an optical signal-to-noise ratio estimator and traffic predictor with Google’s Tensor flow and assessed their performance using experimental and synthetic data Demonstrated the benefits of the proposed cognitive schemes with experiments ona two-domain elastic optical network testbed and evaluations on a six-domain simulation framework Built a two-domain software-defined networking control plane testbed using the ONOS platform and demonstrated the workflow of broker and domain manager cooperation, including hierarchical performance monitoring, topology abstraction and cognitive resource allocation. Designed self-learning cognitive service provisioning schemes using advanced machine learning technologies: Designed a self-taught anomaly detection framework for optical networks, including an unsupervised data clustering module for pattern recognition, a supervised data regression and classification module for online anomaly detection, and the system operation principle Experimentally validated the effectiveness of the self-taught anomaly detection framework with various failure/anomaly scenarios. Designed a self-learning autonomic routing, modulation and spectrum assignment (RMSA) provisioning framework for elastic optical networks with deep reinforcement learning. Designed effective modeling and training mechanisms for the self-learning RMSA framework Conducted extensive numerical simulations with various traffic patterns and topologies to verify the advantage of the self-learning RMSA framework.

42 ENGINEERING↗

Investigating power capping toward energy-efficient scientific applications: Investigating Power Capping toward Energy-Efficient Scientific Applications

The emergence of power efficiency as a primary constraint in processor and system design poses new challenges concerning power and energy awareness for numerical libraries and scientific applications. Power consumption also plays a major role in the design of data centers, which may house petascale or exascale-level computing systems. At these extreme scales, understanding and improving the energy efficiency of numerical libraries and their related applications becomes a crucial part of the successful implementation and operation of the computing system. In this paper, we study and investigate the practice of controlling a compute system's power usage, and we explore how different power caps affect the performance of numerical algorithms with different computational intensities. Further, we determine the impact, in terms of performance and energy usage, that these caps have on a system running scientific applications. This analysis will enable us to characterize the types of algorithms that benefit most from these power management schemes. Our experiments are performed using a set of representative kernels and several popular scientific benchmarks. Lastly, we quantify a number of power and performance measurements and draw observations and conclusions that can be viewed as a roadmap to achieving energy efficiency in the design and execution of scientific algorithms.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Thermosyphon Cooler Hybrid System for Water Savings in an Energy-Efficient HPC Data Center: Results from 24 Months and the Impact on Water Usage Effectiveness

In August 2016, the National Renewable Energy Laboratory (NREL) installed a thermosyphon hybrid cooling system to reduce water usage in its already extremely energy-efficient High-Performance Computing (HPC) Data Center. In its first year of use, the system saved 4,400 m3 (1.16 million gal) of water, and 7,950 m3 (2.10 million gal) during a 2-year period, cutting the use of water in the data center by about one-half. NREL's 930-m2 (10,000-ft2) HPC Data Center is often called the most energy-efficient data center in the world: it has achieved a trailing 12-month average power usage effectiveness of 1.034, and it features a chiller-less design, component-level warm-water liquid cooling, and waste heat capture and reuse. NREL considered the amount of water used by the cooling towers to be counter to the laboratory's sustainability mission, so a team of researchers from NREL, Sandia National Laboratories (Sandia), and Johnson Controls integrated the BlueStream thermosyphon cooler (TSC) - an advanced dry cooler that uses refrigerant in a passive cycle to dissipate heat - on the roof of NREL's Energy Systems Integration Facility, the building that houses the HPC Data Center. In combination with the existing cooling towers, the TSC forms an extremely water- and cost-efficient cooling system. In its first year of operation, on-site water usage effectiveness (WUE) was 0.70 L/kWh. In comparison, the WUE would be 1.27 L/kWh if NREL had continued using only heat-recovery and cooling towers. This on-site water savings was accomplished without negatively impacting the energy-efficient operation of the HPC Data Center. The TSC system technology has the potential for application in data centers around the world, and it is currently being implemented by Sandia. center by about one-half. NREL's 930-m2 (10,000-ft2) HPC Data Center is often called the most energy-efficient data center in the world: it has achieved a trailing 12-month average power usage effectiveness of 1.034, and it features a chiller-less design, component-level warm-water liquid cooling, and waste heat capture and reuse. NREL considered the amount of water used by the cooling towers to be counter to the laboratory's sustainability mission, so a team of researchers from NREL, Sandia National Laboratories (Sandia), and Johnson Controls integrated the BlueStream thermosyphon cooler (TSC) - an advanced dry cooler that uses refrigerant in a passive cycle to dissipate heat - on the roof of NREL's Energy Systems Integration Facility, the building that houses the HPC Data Center. In combination with the existing cooling towers, the TSC forms an extremely water- and cost-efficient cooling system. In its first year of operation, on-site water usage effectiveness (WUE) was 0.70 L/kWh. In comparison, the WUE would be 1.27 L/kWh if NREL had continued using only heat-recovery and cooling towers. This on-site water savings was accomplished without negatively impacting the energy-efficient operation of the HPC Data Center. The TSC system technology has the potential for application in data centers around the world, and it is currently being implemented by Sandia.

97 MATHEMATICS AND COMPUTING↗

Toward integrating high-fidelity CFD approaches in the thermal-hydraulic analysis of turbulent dry cask systems

Nuclear power plants have been supplying resilient and reliable electricity for decades, contributing to energy independence of the U.S.. However, nuclear waste management remains one of the most significant challenges in the industry. The safety of dry cask storage systems relies heavily on their thermal-hydraulic performance. Computational Fluid Dynamics (CFD) simulations are often used to demonstrate this performance and ensure that the system design meets safety standards. This study presents reduced numerical models for various types of dry cask systems. These numerical models can produce efficient and fast results based on the employed modeling strategies. Additionally, the study uses a novel approach to high-fidelity simulations to evaluate modeling assumptions in dry cask modeling. Large Eddy Simulations (LES) are used for this purpose, particularly in regions where fluid velocity is relatively high and the turbulence characteristics become important. Furthermore, the results of these high-fidelity simulations will enhance the interpretation of outcomes produced from a lower-fidelity CFD model.

CFD↗

Requirements specification for nickel cadmium battery expert system

The requirements for performance, design, test, and qualification of a computer program identified as NICBES, Nickel Cadmium Battery Expert System, is established. The specific spacecraft power system configuration selected was the Hubble Space Telescope (HST) Electrical Power System (EPS) Testbed. Power for the HST comes from a system of 13 Solar Panel Arrays (SPAs) linked to 6 Nickel Cadmium Batteries which are connected to 3 Busses. An expert system, NICBES, will be developed at Martin Marietta Aerospace to recognize a testbed anomaly, identify the malfunctioning component and recommend a course of action. Besides fault diagnosis, NICBES will be able to evaluate battery status, give advice on battery status and provide decision support for the operator. These requirements are detailed.

Source record↗

Advanced teleoperation - Progress and problems

A dual-arm Advanced Teleoperation System (ATOPS) was developed at JPL to demonstrate and evaluate performance capabilities of computer and sensing aided teleoperation. This paper gives a brief overview of the system and briefly describes some of the generic and application-oriented performance experiments together with the major results. A major conclusion is that the new advanced technical features of the system should be properly integrated with or interfaced to the natural features of human operations in order to produce an improved overall system and task performance capability.

Bejczy, Antal K.↗

A Collaborative Analysis Tool for Integrated Hypersonic Aerodynamics, Thermal Protection Systems, and RBCC Engine Performance for Single Stage to Orbit Vehicles

Presented is a computer-based tool that connects several disciplines that are needed in the complex and integrated design of high performance reusable single stage to orbit (SSTO) vehicles. Every system is linked to every other system, as is the case of SSTO vehicles with air breathing propulsion, which is currently being studied by NASA. An RBCC propulsion system integrates airbreathing and rocket propulsion into a single engine assembly enclosed within a cowl or duct. A typical RBCC propulsion system operates as a ducted rocket up to approximately Mach 3. Then there is a transition to a ramjet mode for supersonic-to-hypersonic acceleration. Around Mach 8 the engine transitions to a scramjet mode. During the ramjet and scramjet modes, the integral rockets operate as fuel injectors. Around Mach 10-12 (the actual value depends on vehicle and mission requirements), the inlet is physically closed and the engine transitions to an integral rocket mode for orbit insertion. A common feature of RBCC propelled vehicles is the high degree of integration between the propulsion system and airframe. At high speeds the vehicle forebody is fundamentally part of the engine inlet, providing a compression surface for air flowing into the engine. The compressed air is mixed with fuel and burned. The combusted mixture must be expanded to an area larger than the incoming stream to provide thrust. Since a conventional nozzle would be too large, the entire lower after body of the vehicle is used as an expansion surface. Because of the high external temperatures seen during atmospheric flight, the design of an airbreathing SSTO vehicle requires delicate tradeoffs between engine design, vehicle shape, and thermal protection system (TPS) sizing in order to produce an optimum system in terms of weight (and cost) and maximum performance. To adequately determine the performance of the engine/vehicle, the Hypersonic Flight Inlet Model (HYFIM) module was designed to interface with the RBCC engine model. HYFIM performs the aerodynamic analysis of forebodies and inlet characteristics of RBCC powered SSTO launch vehicles. HYFIM is applicable to the analysis of the ramjet/scramjet engine operations modes (Mach 3-12), and provides estimates of parameters such as air capture area, shock-on-lip Mach number, design Mach number, compression ratio, etc., based on a basic geometry routine for modeling axisymmetric cones, 2-D wedge geometries. HYFIM also estimates the variation of shock layer properties normal to the forebody surface. The thermal protection system (TPS) is directly linked to determination of the vehicle moldline and the shaping of the trajectory. Thermal protection systems to maintain the structural integrity of the vehicle must be able to mitigate the heat transfer to the structure and be lightweight. Herein lies the interdependency, in that as the vehicle's speed increases, the TPS requirements are increased. And as TPS masses increase the effect on the propulsion system and all other systems is compounded. The need to analyze vehicle forebody and engine inlet is critical to be able to design the RBCC vehicle. To adequately determine insulation masses for an RBCC vehicle, the hypersonic aerodynamic environment and aeroheating loads must be calculated and the TPS thicknesses must be calculated for the entire vehicle. To accomplish this an ascent or reentry trajectory is obtained using the computer code Program to Optimize Simulated Trajectories (POST). The trajectory is then used to calculate the convective heat rates on several locations on the vehicles using the Miniature Version of the JA70 Aerodynamic Heating Computer Program (MINIVER). Once the heat rates are defined for each body point on the vehicle, then insulation thicknesses that are required to maintain the vehicle within structural limits are calculated using Systems Improved Numerical Differencing Analyzer (SINDA) models. If the TPS masses are too heavy for the performance of the vehicle the process may be repeated altering the trajectory or some other input to reduce the TPS mass. E-PSURBCC is an "engine performance" model and requires the specification of inlet air static temperature and pressure as well as Mach number (which it pulls from the HYFIM and POST trajectory files), and calculates the corresponding stagnation properties. The engine air flow path geometry includes inlet, a constant area section where the rocket is positioned, a subsonic diffuser, a constant area afterburner, and either a converging nozzle or a converging-diverging nozzle. The current capabilities of E-PSURBCC ejector and ramjet mode treatment indicated that various complex flow phenomena including multiple choking and internal shocks can occur for combinations of geometry/flow conditions. For a given input deck defining geometry/flow conditions, the program first goes through a series of checks to establish whether the input parameters are sound in terms of a solution path. If the vehicle/engine performance fails mission goals, the engineer is able to collaboratively alter the vehicle moldline to change aerodynamics, or trajectory, or some other input to achieve orbit. The problem described is an example of the need for collaborative design and analysis. RECIPE is a cross-platform application capable of hosting a number of engineers and designers across the Internet for distributed and collaborative engineering environments. Such integrated system design environments allow for collaborative team design analysis for performing individual or reduced team studies. To facilitate the larger number of potential runs that may need to be made, RECIPE connects the computer codes that calculate the trajectory data, aerodynamic data based on vehicle geometry, heat rate data, TPS masses, and vehicle and engine performance, so that the output from each tool is easily transferred to the model input files that need it.

Stanley, Thomas Troy↗

Application Results on Early Exascale Hardware

This Exascale Computing Project (ECP) milestone report summarizes the status of 27 of the 31 ECP Applications Development (AD) subprojects at the end of FY21. In November and December of 2021, a comprehensive assessment of AD projects was conducted by the ECP leadership along with external subject matter experts (SMEs). (NNSA application projects are reviewed separately using the ASC milestone process.) The AD review committee—consisting of the AD lead, AD deputy, Level 3 (L3), and at least one external project SME—was tasked with evaluating each project’s progress relative to ECP project goals specified in the FY21 timeline. Key areas of focus were code maturity and performance on pre-exascale systems, an in-depth analysis of final key performance parameter (KPP) verification contracts, and future R&D priorities in the final year of ECP and beyond. As such, this report contains not only an accurate snapshot of each subproject’s current status but also represents a broad account of successes and challenges in porting large scientific applications to DOE’s next-generation high-performance computing architectures – the Frontier and Aurora systems.

97 MATHEMATICS AND COMPUTING↗

Mathematical models for space shuttle ground systems

Math models are a series of algorithms, comprised of algebraic equations and Boolean Logic. At Kennedy Space Center, math models for the Space Shuttle Systems are performed utilizing the Honeywell 66/80 digital computers, Modcomp II/45 Minicomputers and special purpose hardware simulators (MicroComputers). The Shuttle Ground Operations Simulator operating system provides the language formats, subroutines, queueing schemes, execution modes and support software to write, maintain and execute the models. The ground systems presented consist primarily of the Liquid Oxygen and Liquid Hydrogen Cryogenic Propellant Systems, as well as liquid oxygen External Tank Gaseous Oxygen Vent Hood/Arm and the Vehicle Assembly Building (VAB) High Bay Cells. The purpose of math modeling is to simulate the ground hardware systems and to provide an environment for testing in a benign mode. This capability allows the engineers to check out application software for loading and launching the vehicle, and to verify the Checkout, Control, & Monitor Subsystem within the Launch Processing System. It is also used to train operators and to predict system response and status in various configurations (normal operations, emergency and contingent operations), including untried configurations or those too dangerous to try under real conditions, i.e., failure modes.

Tory, E. G.↗

NASA HPCC Technology for Aerospace Analysis and Design

The Computational Aerosciences (CAS) Project is part of NASA's High Performance Computing and Communications Program. Its primary goal is to accelerate the availability of high-performance computing technology to the US aerospace community-thus providing the US aerospace community with key tools necessary to reduce design cycle times and increase fidelity in order to improve safety, efficiency and capability of future aerospace vehicles. A complementary goal is to hasten the emergence of a viable commercial market within the aerospace community for the advantage of the domestic computer hardware and software industry. The CAS Project selects representative aerospace problems (especially design) and uses them to focus efforts on advancing aerospace algorithms and applications, systems software, and computing machinery to demonstrate vast improvements in system performance and capability over the life of the program. Recent demonstrations have served to assess the benefits of possible performance improvements while reducing the risk of adopting high-performance computing technology. This talk will discuss past accomplishments in providing technology to the aerospace community, present efforts, and future goals. For example, the times to do full combustor and compressor simulations (of aircraft engines) have been reduced by factors of 320:1 and 400:1 respectively. While this has enabled new capabilities in engine simulation, the goal of an overnight, dynamic, multi-disciplinary, 3-dimensional simulation of an aircraft engine is still years away and will require new generations of high-end technology.

Schulbach, Catherine H.↗

Phase-field modeling of diffusional phase behaviors of solid surfaces: A case study of phase-separating Li X FePO 4 electrode particles

In this paper, we present a comprehensive phase-field model for simulating diffusion-mediated kinetic phase behaviors near the surface of a solid particle. The model incorporates elastic inhomogeneity and anisotropy, diffusion mobility anisotropy, interfacial energy anisotropy, and Cahn–Hilliard diffusion kinetics. The free energy density function is formulated based on the regular solution model taking into account the possible solute-surface interaction near the surface. The coherency strain energy is computed using the Fourier-spectral iterative-perturbation method due to the strong elastic inhomogeneity with a zero surface traction boundary condition. Employing a phase-separating Li X FePO 4 electrode particle for Li-ion batteries as a model system, we perform parametric three-dimensional computer simulations. The model permits the observation of surface phase behaviors that are different from the bulk counterpart. For instance, it reproduces the theoretically well-established surface modes of spinodal decomposition of an unstable solid solution: the surface mode of coherent spinodal decomposition and the surface-directed spinodal decomposition mode. We systematically investigate the influences of major factors on the kinetic surface phase behaviors during the diffusional process. Finally, our simulation study provides insights for tailoring the internal phase microstructure of a particle by controlling the surface phase morphology.

25 ENERGY STORAGE↗

PaRSEC: Scalability, flexibility, and hybrid architecture support for task-based applications in ECP

This paper highlights the most significant enhancements made to PaRSEC, a scalable task-based runtime system designed for hybrid machines, during the Exascale Computing Project (ECP). The enhancements focus on expanding the capabilities of PaRSEC to address the evolving landscape of parallel computing. Notable achievements include the integration of support for three major types of accelerators (NVIDIA, AMD, and Intel GPUs), the refinement and increased flexibility of the communication subsystem, and the introduction of new programming interfaces tailored for irregular applications. Additionally, the project resulted in the development of powerful debugging and performance analysis tools aimed at assisting users in understanding and optimizing their applications. We present a comprehensive demonstration of these advancements through a series of benchmarks and applications within ECP and beyond, thereby showcasing the enhanced capabilities of PaRSEC across the diverse architectures within the ECP, providing valuable insights into the runtime system’s adaptability and performance across varied computing environments.

Bouteiller, Aurelien↗

Climate Analytics as a Service

Exascale computing, big data, and cloud computing are driving the evolution of large-scale information systems toward a model of data-proximal analysis. In response, we are developing a concept of climate analytics as a service (CAaaS) that represents a convergence of data analytics and archive management. With this approach, high-performance compute-storage implemented as an analytic system is part of a dynamic archive comprising both static and computationally realized objects. It is a system whose capabilities are framed as behaviors over a static data collection, but where queries cause results to be created, not found and retrieved. Those results can be the product of a complex analysis, but, importantly, they also can be tailored responses to the simplest of requests. NASA's MERRA Analytic Service and associated Climate Data Services API provide a real-world example of climate analytics delivered as a service in this way. Our experiences reveal several advantages to this approach, not the least of which is orders-of-magnitude time reduction in the data assembly task common to many scientific workflows.

big data↗

The formulation of simulation models of aeronautical systems

Recent developments in formula-manipulation compilers have led to a significant reduction in the time spent in formulating models of complex multi-degrees-of-freedom dynamical systems. MACSYMA, a computer system that can be used to perform symbolic manipulations in an interactive mode, is applied to the problem of formulating models of aeronautical systems for simulation studies. An example demonstrates that once the procedure is established, the formulation and modification of the models can be reduced to a series of routine computer operations.

Howard, J. C.↗

Design Space Exploration of Emerging Memory Technologies for Machine Learning Applications

Memory design space exploration methods study memory systems’ performances and limitations before implementation. The computer memory design space has grown exponentially because of the enormous growth of memory types, memory controllers, and application software. Computer simulators are commonly used for memory design space exploration. However, complex memory simulations take an enormous amount of time. Hence, in this paper, we proposed a machine learning-based design space exploration method for dynamic random-access memory and non-volatile memory systems. We applied our method to the CosmoGAN and LeNet applications to predict the following six memory response parameters: (i) bandwidth, (ii) power, (iii) average latency, (iv) average total latency, (v) memory reads, and (vi) memory writes. Our experimental results show that machine learning models can predict memory response parameter values faster than simulations. We used support vector machine, random forest, and gradient boosting machine learning models. We observed that the support vector machine provides better performance for bandwidth, average latency, and average total latency. The random forest model works better for memory reads and writes. The gradient boosting model provides superior prediction performance for power. We provide a detailed discussion on learning curve characteristics, error analysis, and memory type recommendation.

Hasan, S M Shamimul↗

Physics-Based Optical Neuromorphic Classification

Typical approaches to classify scenes from light convert the light field to electrons to perform the computation in the digital electronic domain. This conversion and downstream computational analysis require significant power and time. Diffractive neural networks have recently emerged as unique systems to classify optical fields at lower energy and high speeds. Previous work has shown that a single layer of diffractive metamaterial can achieve high performance on classification tasks. In analogy with electronic neural networks, it is anticipated that multilayer diffractive systems would provide better performance, but the fundamental reasons for the potential improvement have not been established. In this work, we present extensive computational simulations of two - layer diffractive neural networks and show that they can achieve high performance with fewer diffractive features than single layer systems.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗