SEARCH · Search NASA

Results for “fault-tolerant”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 145 records · Page 8

All sources NASA NTRS DOE OSTI

NASA NTRS1983

Operational characteristics of the dispersed sensor processor mesh

The dispersed sensor processing mesh (DSPM) is part of an experimental system used to pursue a proof of concept for an advanced fault-tolerant communication strategy. The DSPM experiments incorporate sensors and effectors into an ultra-reliable nodal network in which a central bus controller performs algorithms to grow and maintain the network. The experiments demonstrate that DSPM is an achievable communication network that can sustain failures and continue to function properly. The results of the experiment give insights into the unique characteristics of the DSPM network and the network's capacity to limit physical damage propagation, limit damaged-data propagation, and to survive cyclic communication paths. This paper describes the concept, mechanization, and performance of the DSPM system in a real-time flight-critical environment.

Abbott, L. W.↗

NASA NTRS1983

Fault detection, isolation and reconfiguration in FTMP Methods and experimental results

The Fault-Tolerant Multiprocessor (FTMP) is a highly reliable computer designed to meet a goal of 10 to the -10th failures per hour and built with the objective of flying an active-control transport aircraft. Fault detection, identification, and recovery software is described, and experimental results obtained by injecting faults in the pin level in the FTMP are presented. Over 21,000 faults were injected in the CPU, memory, bus interface circuits, and error detection, masking, and error reporting circuits of one LRU of the multiprocessor. Detection, isolation, and reconfiguration times were recorded for each fault, and the results were found to agree well with earlier assumptions made in reliability modeling.

Lala, J. H.↗

NASA NTRS1984

Space station - Technology development

The NASA manned space station program's systems technology effort involves the development of novel techniques that will reduce the scope of tasks neeeded for design, development, testing and evaluation of the hardware. Operations technology efforts encompass analyses that will define those techniques best able to improve the efficiency and reduce the costs of space station functions. The technology objective for data management calls for a fault-tolerant, distributed, expandable and adaptable, as well as repairable and user-friendly, flight data management system that employs state-of-the-art hardware and software. The space station's power system includes the largest element, a 'solar blanket', and the heaviest component, the batteries, of all the subsystems. A thermal management system for the power system is of paramount importance. Attention is also given to the exacting demands of attitude control and stabilization and a regenerative life support system of the requisite capacity and reliability.

Carlisle, R. F.↗

NASA NTRS1984

Test experience on an ultrareliable computer communication network

The dispersed sensor processing mesh (DSPM) is an experimental, ultra-reliable, fault-tolerant computer communications network that exhibits an organic-like ability to regenerate itself after suffering damage. The regeneration is accomplished by two routines - grow and repair. This paper discusses the DSPM concept for achieving fault tolerance and provides a brief description of the mechanization of both the experiment and the six-node experimental network. The main topic of this paper is the system performance of the growth algorithm contained in the grow routine. The characteristics imbued to DSPM by the growth algorithm are also discussed. Data from an experimental DSPM network and software simulation of larger DSPM-type networks are used to examine the inherent limitation on growth time by the growth algorithm and the relationship of growth time to network size and topology.

Abbott, L. W.↗

NASA NTRS1984

Data and results of a laboratory investigation of microprocessor upset caused by simulated lightning-induced analog transients

Advanced composite aircraft designs include fault-tolerant computer-based digital control systems with thigh reliability requirements for adverse as well as optimum operating environments. Since aircraft penetrate intense electromagnetic fields during thunderstorms, onboard computer systems maya be subjected to field-induced transient voltages and currents resulting in functional error modes which are collectively referred to as digital system upset. A methodology was developed for assessing the upset susceptibility of a computer system onboard an aircraft flying through a lightning environment. Upset error modes in a general-purpose microprocessor were studied via tests which involved the random input of analog transients which model lightning-induced signals onto interface lines of an 8080-based microcomputer from which upset error data were recorded. The application of Markov modeling to upset susceptibility estimation is discussed and a stochastic model development.

Belcastro, C. M.↗

NASA NTRS1985

Care 3 model overview and user's guide, first revision

A manual was written to introduce the CARE III (Computer-Aided Reliability Estimation) capability to reliability and design engineers who are interested in predicting the reliability of highly reliable fault-tolerant systems. It was also structured to serve as a quick-look reference manual for more experienced users. The guide covers CARE III modeling and reliability predictions for execution in the CDC CYber 170 series computers, DEC VAX-11/700 series computer, and most machines that compile ANSI Standard FORTRAN 77.

Bavuso, S. J.↗

NASA NTRS1983

Peer Review of a Formal Verification/Design Proof Methodology

The role of formal verification techniques in system validation was examined. The value and the state of the art of performance proving for fault-tolerant compuers were assessed. The investigation, development, and evaluation of performance proving tools were reviewed. The technical issues related to proof methodologies are examined. The technical issues discussed are summarized.

Source record↗

NASA NTRS1983

Some methods of estimating a coverage parameter

To demonstrate the high reliability of fault-tolerant computing systems, experimental testing can be performed by injecting faults into their hardware and observing the times needed to detect and isolate failed components and reconfigure the good components. Coverage, a parameter measuring continued system success, is the probability that recovery occurs before other, perhaps catastrophic, component failures occur. Methods of estimating coverage are given for the case of randomly selecting a subset of the pins or chips for testing. Pin-level samples of times to recovery are treated by the Type I censoring model often used in life testing.

Lee, L. D.↗

NASA NTRS1984

Test experience on an ultrareliable computer communication network

The dispersed sensor processing mesh (DSPM) is an experimental, ultrareliable, fault-tolerant computer communications network that exhibits an organic-like ability to regenerate itself after suffering damage. The regeneration is accomplished by two routines - grow and repair. This paper discusses the DSPM concept for achieving fault tolerance and provides a brief description of the mechanization of both the experiment and the six-node experimental network. The main topic of this paper is the system performance of the growth algorithm contained in the grow routine. The characteristics imbued to DSPM by the growth algorithm are also discussed. Data from an experimental DSPM network and software simulation of larger DSPM-type networks are used to examine the inherent limitation on growth time by the growth algorithm and the relationship of growth time to network size and topology.

Abbott, L. W.↗

NASA NTRS1984

Autonomous spacecraft design methodology

A methodology for autonomous spacecraft design blends autonomy requirements with traditional mission requirements and assesses the impact of autonomy upon the total system resources available to support fault-tolerance and automation. A baseline functional design can be examined for autonomy implementation impacts, and the costs, risk, and benefits of various options can be assessed. The result of the process is a baseline design that includes autonomous control functions.

Divita, E. L.↗

NASA NTRS1984

Validation of fault-free behavior of a reliable multiprocessor system - FTMP: A case study

A program of experiments has been conducted at NASA-Langley to test the fault-free performance of a Fault-Tolerant Multiprocessor (FTMP) avionics system for next-generation aircraft. Baseline measurements of an operating FTMP system were obtained with respect to the following parameters: instruction execution time, frame size, and the variation of clock ticks. The mechanisms of frame stretching were also investigated. The experimental results are summarized in a table. Areas of interest for future tests are identified, with emphasis given to the implementation of a synthetic workload generation mechanism on FTMP.

Clune, E.↗

NASA NTRS1984

A user's view of CARE III

The present computerized reliability predictor for digital fault-tolerant systems whose sizes are of the order of one million Markovian equivalent states, employs advanced stochastic modeling techniques and implements a mixed Markov model that enables it to drastically reduce the state size of hitherto computationally unobtainable models. Attention is given to the concepts of failure, fault, and error, in the context of the novel system's fault/error-handling models. Examples are drawn from the system's user-friendly interface dialog.

Bavuso, S. J.↗

NASA NTRS1984

Automated ultrareliability models - A review

Analytic models are required to assess the reliability of systems designed to ultrareliability requirements. This paper reviews the capabilities and limitations of five currently available automated reliability models which are applicable to fault-tolerant flight control systems. 'System' includes sensors, computers, and actuators. A set of review criteria including validation, configuration adaptability, and resource requirements for model evaluation are described. Five models, ARIES, CARE II, CARE III, CARSRA, and CAST, are assessed against the criteria, thereby characterizing their capabilities and limitations. This review should be helpful to potential users of the models.

Bridgman, M. S.↗

NASA NTRS1985

Fault-free behavior of reliable multiprocessor systems: FTMP experiments in AIRLAB

This report describes a set of experiments which were implemented on the Fault tolerant Multi-Processor (FTMP) at NASA/Langley's AIRLAB facility. These experiments are part of an effort to formulate and evaluate validation methodologies for fault-tolerant computers. This report deals with the measurement of single parameters (baselines) of a fault free system. The initial set of baseline experiments lead to the following conclusions: (1) The system clock is constant and independent of workload in the tested cases; (2) the instruction execution times are constant; (3) the R4 frame size is 40mS with some variation; (4) the frame stretching mechanism has some flaws in its implementation that allow the possibility of an infinite stretching of frame duration. Future experiments are planned. Some will broaden the results of these initial experiments. Others will measure the system more dynamically. The implementation of a synthetic workload generation mechanism for FTMP is planned to enhance the experimental environment of the system.

Clune, E.↗

NASA NTRS1985

The embedded operating system project

The design and construction of embedded operating systems for real-time advanced aerospace applications was investigated. The applications require reliable operating system support that must accommodate computer networks. Problems that arise in the construction of such operating systems, reconfiguration, consistency and recovery in a distributed system, and the issues of real-time processing are reported. A thesis that provides theoretical foundations for the use of atomic actions to support fault tolerance and data consistency in real-time object-based system is included. The following items are addressed: (1) atomic actions and fault-tolerance issues; (2) operating system structure; (3) program development; (4) a reliable compiler for path Pascal; and (5) mediators, a mechanism for scheduling distributed system processes.

Campbell, R. H.↗

NASA NTRS1985

Integrated analysis of error detection and recovery

An integrated modeling and analysis of error detection and recovery is presented. When fault latency and/or error latency exist, the system may suffer from multiple faults or error propagations which seriously deteriorate the fault-tolerant capability. Several detection models that enable analysis of the effect of detection mechanisms on the subsequent error handling operations and the overall system reliability were developed. Following detection of the faulty unit and reconfiguration of the system, the contaminated processes or tasks have to be recovered. The strategies of error recovery employed depend on the detection mechanisms and the available redundancy. Several recovery methods including the rollback recovery are considered. The recovery overhead is evaluated as an index of the capabilities of the detection and reconfiguration mechanisms.

Shin, K. G.↗

NASA NTRS1985

Achieving reliability - The evolution of redundancy in American manned spacecraft computers

The Shuttle is the first launch system deployed by NASA with full redundancy in the on-board computer systems. Fault-tolerance, i.e., restoring to a backup with less capabilities, was the method selected for Apollo. The Gemini capsule was the first to carry a computer, which also served as backup for Titan launch vehicle guidance. Failure of the Gemini computer resulted in manual control of the spacecraft. The Apollo system served vehicle flight control and navigation functions. The redundant computer on Skylab provided attitude control only in support of solar telescope pointing. The STS digital, fly-by-wire avionics system requires 100 percent reliability. The Orbiter carries five general purpose computers, four being fully-redundant and the fifth being soley an ascent-descent tool. The computers are synchronized at input and output points at a rate of about six times a second. The system is projected to cause a loss of an Orbiter only four times in a billion flights.

Tomayko, J. E.↗

NASA NTRS1985

A theoretical basis for the analysis of multiversion software subject to coincident errors

Fundamental to the development of redundant software techniques (known as fault-tolerant software) is an understanding of the impact of multiple joint occurrences of errors, referred to here as coincident errors. A theoretical basis for the study of redundant software is developed which: (1) provides a probabilistic framework for empirically evaluating the effectiveness of a general multiversion strategy when component versions are subject to coincident errors, and (2) permits an analytical study of the effects of these errors. An intensity function, called the intensity of coincident errors, has a central role in this analysis. This function describes the propensity of programmers to introduce design faults in such a way that software components fail together when executing in the application environment. A condition under which a multiversion system is a better strategy than relying on a single version is given.

Eckhardt, D. E., Jr.↗