Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Computer Systems”

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 1,621 records · Page 90
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Comparative evaluation of deep learning workloads for leadership-class systems

Deep learning (DL) workloads and their performance at scale are becoming important factors to consider as we design, develop and deploy next-generation high-performance computing systems. Since DL applications rely heavily on DL frameworks and underlying compute (CPU/GPU) stacks, it is essential to gain a holistic understanding from compute kernels, models, and frameworks of popular DL stacks, and to assess their impact on science-driven, mission-critical applications. At Oak Ridge Leadership Computing Facility (OLCF), we employ a set of micro and macro DL benchmarks established through the Collaboration of Oak Ridge, Argonne, and Livermore (CORAL) to evaluate the AI readiness of our next-generation supercomputers. In this paper, we present our early observations and performance benchmark comparisons between the Nvidia V100 based Summit system with its CUDA stack and an AMD MI100 based testbed system with its ROCm stack. We take a layered perspective on DL benchmarking and point to opportunities for future optimizations in the technologies that we consider.

Yin, Junqi↗
NASA NTRS2021

A Simulation Study of Bin-and-Sort Policies in a Distributed System for Flights Scheduling

This is a report on the results of a simulation experiment conducted on an automated distributed system for scheduling flights in a shared airspace. The scheduling is carried out by computer systems owned by the private operators of the flights, which interface through a centralized independent server. The scheduling requires determining which of the group of flights at hand (whether owned by the same operator or different ones) are to have the advantage of being scheduled before the others. A chosen method of grouping and prioritizing flights must: have the ability to schedule a flight promptly, prevent scheduling conflicts between pairs of overlapping flight routes, and be computationally feasible. Scheduling flights one by one is the First Come First Served policy. Scheduling too many at a time is infeasible. The type of scheduling policy proposed herein, called bin-and-sort, is an attempt to reach a compromise between these two extremes. Bin-and-sort allows for each operator to choose its criteria for prioritizing its flights, and for the centralized component to use different arbitration criteria. A number of these criteria are used in the simulation presented herein. The results of each criterion choice are reported and analyzed.

flight scheduling↗
NASA NTRS2021

A Simulation Study of Bin-and-Sort Policies in a Distributed System for Flights Scheduling

This is a report on the results of a simulation experiment conducted on an automated distributed system for scheduling flights in a shared airspace. The scheduling is carried out by computer systems owned by the private operators of the flights, which interface through a centralized independent server. The scheduling requires determining which of the group of flights at hand (whether owned by the same operator or different ones) are to have the advantage of being scheduled before the others. A chosen method of grouping and prioritizing flights must: have the ability to schedule a flight promptly, prevent scheduling conflicts between pairs of overlapping flight routes, and be computationally feasible. Scheduling flights one by one is the First Come First Served policy. Scheduling too many at a time is infeasible. The type of scheduling policy proposed herein, called bin-and-sort, is an attempt to reach a compromise between these two extremes. Bin-and-sort allows for each operator to choose its criteria for prioritizing its flights, and for the centralized component to use different arbitration criteria. A number of these criteria are used in the simulation presented herein. The results of each criterion choice are reported and analyzed.

flight scheduling↗
DOE OSTI2019

Memristors learn to play

A hybrid analogue–digital computing system built upon memristive devices is capable of solving classic control problems with potentially a lower energy consumption and higher speed than fully digital systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

Foundational Tools for Petascale Computing

The Paradyn project has a history of developing algorithms, techniques, and software that push the cutting edge of tool technology for high-end computing systems. Under this funding, we are working on a three-year agenda to make substantial new advances in support of new and emerging Petascale systems. The overall goal for this work is to address the steady increase in complexity of these petascale systems. Our work covers two key areas: (1) The analysis, instrumentation and control of binary programs. Work in this area falls under the general framework of the Dyninst API tool kits. (2) Infrastructure for building tools and applications at extreme scale. Work in this area falls under the general framework of the MRNet scalability framework. Note that work done under this funding is closely related to work done under a contemporaneous grant, “High-Performance Energy Applications and Systems”, SC0004061/FG02-10ER25972, UW PRJ36WV.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Systems Security Architecture Research and Development

The general technical goals of the project were to better understand the assumptions of security guarantees using secure commercial off-the-shelf (COTS) technologies, to identify possible security weaknesses introduced during system integration, and to create improved security hardware and software which result in well understood reference architectures for critical computing systems and applications.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2003

Toward a Dynamically Reconfigurable Computing and Communication System for Small Spacecraft

Future science missions will require the use of multiple spacecraft with multiple sensor nodes autonomously responding and adapting to a dynamically changing space environment. The acquisition of random scientific events will require rapidly changing network topologies, distributed processing power, and a dynamic resource management strategy. Optimum utilization and configuration of spacecraft communications and navigation resources will be critical in meeting the demand of these stringent mission requirements. There are two important trends to follow with respect to NASA's (National Aeronautics and Space Administration) future scientific missions: the use of multiple satellite systems and the development of an integrated space communications network. Reconfigurable computing and communication systems may enable versatile adaptation of a spacecraft system's resources by dynamic allocation of the processor hardware to perform new operations or to maintain functionality due to malfunctions or hardware faults. Advancements in FPGA (Field Programmable Gate Array) technology make it possible to incorporate major communication and network functionalities in FPGA chips and provide the basis for a dynamically reconfigurable communication system. Advantages of higher computation speeds and accuracy are envisioned with tremendous hardware flexibility to ensure maximum survivability of future science mission spacecraft. This paper discusses the requirements, enabling technologies, and challenges associated with dynamically reconfigurable space communications systems.

Kifle, Muli↗
DOE OSTI2021

Full-field imaging learning machine (FILM)

A method of determining dynamic properties of a structure (linear or nonlinear) includes receiving spatio-temporal inputs, generating mode shapes and modal components corresponding to the spatio-temporal inputs using a trained deep complexity coding artificial neural network, and subsequently generating the dynamic properties by analyzing each modal component using a trained learning machine. A computing system for non-contact determination of dynamic properties of a structure includes a camera, a processor, and a memory including computer-executable instructions. When the instructions are executed, the system is caused to receive spatio-temporal image data, decompose the spatio-temporal image data into constituent manifold components using an autoencoder, and analyze the constituent manifold components using a trained learning machine to determine the dynamic properties.

Yang, Yongchao↗
DOE OSTI2020

Power-oriented bus encoding for data transmission

Systems, apparatuses, and methods for reducing the toggle rates on buses are disclosed. A computing system includes a source which provides packets for transmission on a bus. The packet is compressed by a compression engine. The compressed data format of the packet includes locations (bit positions) referred to as holes which do not include valid data. A bus configuration module identifies the locations of the holes and replaces the holes with information from a previous packet transmitted earlier on the bus. The bus configuration module also determines a new transmission bus width for the packet for lowering the bus toggle rate on the bus during transmission.

42 ENGINEERING↗
DOE OSTI2024

DIRAC current, upcoming and planned capabilities and technologies

DIRAC is the interware for building and operating large scale distributed computing systems. It is adopted by multiple collaborations from various scientific domains for implementing their computing models. DIRAC provides a framework and a rich set of ready-to-use services for Workload, Data and Production Management tasks of small, medium and large scientific communities having different computing requirements. The base functionality can be easily extended by custom components supporting community specific workflows. DIRAC is at the same time an aging project, and a new DiracX project is taking shape for replacing DIRAC in the long term. This contribution will highlight DIRAC’s current, upcoming and planned capabilities and technologies, and how the transition to DiracX will take place. Examples include, but are not limited to, adoption of security tokens and interactions with Identity Provider services, integration of Clouds and High Performance Computers, interface with Rucio, improved monitoring and deployment procedures.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

Visual Data-Analytics of Large-Scale Parallel Discrete-Event Simulations

Parallel discrete-event simulation (PDES) is an important tool in the codesign of extreme-scale systems because PDES provides a cost-effective way to evaluate designs of highperformance computing systems. Optimistic synchronization algorithms for PDES, such as Time Warp, allow events to be processed without global synchronization among the processing elements. A rollback mechanism is provided when events are processed out of timestamp order. Although optimistic synchronization protocols enable the scalability of large-scale PDES, the performance of the simulations must be tuned to reduce the number of rollbacks and provide an improved simulation runtime. To enable efficient large-scale optimistic simulations, one has to gain insight into the factors that affect the rollback behavior and simulation performance. We developed a tool for ROSS model developers that gives them detailed metrics on the performance of their large-scale optimistic simulations at varying levels of simulation granularity. Model developers can use this information for parameter tuning of optimistic simulations in order to achieve better runtime and fewer rollbacks. In this work, we instrument the ROSS optimistic PDES framework to gather detailed statistics about the simulation engine. We have also developed an interactive visualization interface that uses the data collected by the ROSS instrumentation to understand the underlying behavior of the simulation engine. The interface connects real time to virtual time in the simulation and provides the ability to view simulation data at different granularities. We demonstrate the usefulness of our framework by performing a visual analysis of the dragonfly network topology model provided by the CODES simulation framework built on top of ROSS. The instrumentation needs to minimize overhead in order to accurately collect data about the simulation performance. To ensure that the instrumentation does not introduce unnecessary overhead, we perform a scaling study that compares instrumented ROSS simulations with their noninstrumented counterparts in order to determine the amount of perturbation when running at different simulation scales.

Ross, Caitlin↗
DOE OSTI2019

Quantum Networks for Open Science (QNOS) Workshop

Quantum computing systems currently being developed will have extraordinary capabilities to effectively solve complex problems in computational sciences, communication networks, artificial intelligence, and data processing, and will provide a powerful capability for researchers in almost every scientific discipline. Harnessing the full potential of quantum computing will require an ecosystem with a broad spectrum of quantum technologies. Quantum networks are one of the critical and highly anticipated components of this ecosystem. The combination of quantum computing and quantum networks are crucial to the US Department of Energy’s (DOE) mission to provide scientists with the state-of-the-art computational capabilities.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1972

Feasibility study: Replacement of the inoperative decommutating buffer subsystem for the instrumentation checkout complex in the Quality and Reliability Assurance Laboratory

A general purpose computer system, that is necessary for replacement of the present inoperative signal decommutator special purpose computer subsystem is described. The present decommutator subsystem has a very poor history of reliability and since April 1970, it has become inoperative because the core memory cannot be repaired. Functions of the present signal, decommutator subsystem are to receive, demultiplex, record in real time, playback in real time, and output to the SDS-930 control computer for analysis of the telemetry data. Recommendations for replacement of the inoperative telemetry decommutator subsystem are for the purchase of a mini-computer.

Hilliard, J. W.↗
NASA NTRS1976

Computer-aided design and computer science technology

A description is presented of computer-aided design requirements and the resulting computer science advances needed to support aerospace design. The aerospace design environment is examined, taking into account problems of data handling and aspects of computer hardware and software. The interactive terminal is normally the primary interface between the computer system and the engineering designer. Attention is given to user aids, interactive design, interactive computations, the characteristics of design information, data management requirements, hardware advancements, and computer science developments.

Fulton, R. E.↗
NASA NTRS1973

Feasibility study of an Integrated Program for Aerospace vehicle Design (IPAD). Volume 6: IPAD system development and operation

The strategy of the IPAD implementation plan presented, proposes a three phase development of the IPAD system and technical modules, and the transfer of this capability from the development environment to the aerospace vehicle design environment. The system and technical module capabilities for each phase of development are described. The system and technical module programming languages are recommended as well as the initial host computer system hardware and operating system. The cost of developing the IPAD technology is estimated. A schedule displaying the flowtime required for each development task is given. A PERT chart gives the developmental relationships of each of the tasks and an estimate of the operational cost of the IPAD system is offered.

Redhed, D. D.↗
NASA NTRS1980

Development and analysis of the learjet 54/55 fuselage NASTRAN model using substructure techniques

Development and analysis of the Learjet 54/55 fuselage NASTRAN model is presented. This structure was broken down into several substructures to make the modeling and analysis effort more manageable. Since the geometry was fairly complex in some areas and in order to provide flexibility for future configuration studies, a series of local coordinate systems were used to describe the model. This work was accomplished using several different computer systems, the more recent of which were connected by a high speed data communications lines to perform the tasks of model generation and NASTRAN analysis.

Boroughs, R. R.↗