SEARCH · Search NASA
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.
A survey and analysis of the existing methods, tools and techniques employed in the development of software are presented along with recommendations for the construction of reliable software. Functional designs for software specification language, and the data base verifier are presented.
This case study analyzes the application of the cleanroom software development methodology to the development of production software at the NASA/Goddard Space Flight Center. The cleanroom methodology emphasizes human discipline in program verification to produce reliable software products that are right the first time. Preliminary analysis of the cleanroom case study shows that the method can be applied successfully in the FDD environment and may increase staff productivity and product quality. Compared to typical Software Engineering Laboratory (SEL) activities, there is evidence of lower failure rates, a more complete and consistent set of inline code documentation, a different distribution of phase effort activity, and a different growth profile in terms of lines of code developed. The major goals of the study were to: (1) assess the process used in the SEL cleanroom model with respect to team structure, team activities, and effort distribution; (2) analyze the products of the SEL cleanroom model and determine the impact on measures of interest, including reliability, productivity, overall life-cycle cost, and software quality; and (3) analyze the residual products in the application of the SEL cleanroom model, such as fault distribution, error characteristics, system growth, and computer usage.
Formal specification and analysis of requirements continues to gain support as a method for producing more reliable software. However, the introduction of formal methods to a large software project is difficult, due in part to the unfamiliarity of the specification languages and the lack of graphics. This paper reports results of an investigation into the effectiveness of formal methods as an aid to the requirements analysis of critical, system-level fault-protection software on a spacecraft currently under development. Our experience indicates that formal specification and analysis can enhance the accuracy of the requirements and add assurance prior to design development in this domain. The work described here is part of a larger, NASA-funded research project whose purpose is to use formal-methods techniques to improve the quality of software in space applications. The demonstration project described here is part of the effort to evaluate experimentally the effectiveness of supplementing traditional engineering approaches to requirements specification with the more rigorous specification and analysis available with formal methods.
Formal specification and analysis of requirements continues to gain support as a method for producing more reliable software. However, the introduction of formal methods to a large software project is difficult, due in part to the unfamiliarity of the specification languages and the lack of graphics. This paper reports results of an investigation into the effectiveness of formal methods as an aid to the requirements analysis of critical, system-level fault-protection software on a spacecraft currently under development. Our experience indicates that formal specification and analysis can enhance the accuracy of the requirements and add assurance prior to design development in this domain.
This slide presentation reviews a risk assessment model as applied to software development. the presentation uses graphs to demonstrate basic concepts of software reliability. It also discusses the application to the risk model to the software development life cycle.
Software is becoming increasingly significant in today's critical avionics systems. To achieve safe, reliable software, government regulatory agencies such as the Federal Aviation Administration (FAA) and the Department of Defense mandate the use of certain software development methods. However, little scientific evidence exists to show a correlation between software development methods and product quality. Given this lack of evidence, a series of experiments has been conducted to understand why and how software fails. The Guidance and Control Software (GCS) project is the latest in this series. The GCS project is a case study of the Requirements and Technical Concepts for Aviation RTCA/DO-178B guidelines, Software Considerations in Airborne Systems and Equipment Certification. All civil transport airframe and equipment vendors are expected to comply with these guidelines in building systems to be certified by the FAA for use in commercial aircraft. For the case study, two implementations of a guidance and control application were developed to comply with the DO-178B guidelines for Level A (critical) software. The development included the requirements, design, coding, verification, configuration management, and quality assurance processes. This paper discusses the details of the GCS project and presents the results of the case study.
Report describes radar data system at NASA Western Aeronautical Test Range. System provides real-time and recorded data about flightpaths of research aircraft and Space Shuttle. Called RADATS, processes data from three radars simultaneously; interacts with system operator; enhances data by introducing corrections and smoothing; controls range, azimuth, and elevation of radars; and automatically calibrates itself before and after missions. Software classified into three kinds of programs: utility, real-time, and calibration. Equipment exhibited exceptional reliability. Software matured to become virtually trouble-free.
This paper presents an analysis of operating system failures on an IBM 3081 running VM/SP. Three broad categories of software failures are found: error handling, program control or logic, and hardware related; it is found that more than 25 percent of software failures occur in the hardware/software interface. Measurements show that results on software reliability cannot be considered representative unless the system workload is taken into account. The overall CPU execution rate, although measured to be close to 100 percent most of the time, is not found to correlate strongly with the occurrence of failures. Possible reasons for the observed workload failure dependency, based on detailed investigations of the failure data, are discussed.
A committee of experienced managers representing each major NASA organization guides a major development effort which seeks to ensure the acquisition of economical and reliable software for more complex future programs. The primary, near-term development objectives of the NASA Software Management and Assurance Program are: (1) to provide the best management and technical guidance available; (2) to facilitate the use of proven tools, techniques, and information; and (3) to maintain a pool of highly qualified software personnel. The software development tasks addressed involve such problems as satellite end-to-end architecture and advanced aircraft guidance and control systems.
Products of this project will significantly improve the quality and productivity of Space Station Freedom Program software processes by: improving software reliability and safety; and broadening the range of problems that can be solved with computational solutions. Projects brings in Computer Aided Software Engineering (CASE) technology for: Environments such as Engineering Script Language/Parts Composition System (ESL/PCS) application generator, Intelligent User Interface for cost avoidance in setting up operational computer runs, Framework programmable platform for defining process and software development work flow control, Process for bringing CASE technology into an organization's culture, and CLIPS/CLIPS Ada language for developing expert systems; and methodologies such as Method for developing fault tolerant, distributed systems and a method for developing systems for common sense reasoning and for solving expert systems problems when only approximate truths are known.
Knowledge-based expert systems are playing an increasingly important role in NASA space and aircraft systems. However, many of NASA's software applications are life- or mission-critical and knowledge-based systems do not lend themselves to the traditional verification and validation techniques for highly reliable software. Rule-based systems lack the control abstractions found in procedural languages. Hence, it is difficult to verify or maintain such systems. Our goal is to automatically structure a rule-based system into a set of rule-groups having a well-defined interface to other rule-groups. Once a rule base is decomposed into such 'firewalled' units, studying the interactions between rules would become more tractable. Verification-aid tools can then be developed to test the behavior of each such rule-group. Furthermore, the interactions between rule-groups can be studied in a manner similar to integration testing. Such efforts will go a long way towards increasing our confidence in the expert-system software. Our research efforts address the feasibility of automating the identification of rule groups, in order to decompose the rule base into a number of meaningful units.
The work reported here provides protection against software failures in the task dispatcher of the FTMP, a particularly critical portion of the system software. Faults in other system modules and application programs can be handled by similar techniques but are not covered in this effort. Goals of the work reported here are: (1) to develop provisions in the software design that will detect and mitigate software failures in the dispatcher portion of the FTMP Executive and, (2) to propose the implementation of specific software reliability measures in other parts of the system. Beyond the specific support to the FTMP project, the work reported here represents a considerable advance in the practical application of the recovery block methodology for fault tolerant software design.
In connection with advances in technology, mainly in the electronic area, aircraft flight control applications have evolved from simple pilot-relief autopilots to flight-critical and redundant fly-by-wire and active control systems. For flight-critical implementations which required accommodation of inflight failures, additional levels of redundancy were incorporated to provide fail-safe and fail-operative performance. The current status of flight control systems reliability is examined and high-reliability approaches are discussed. Attention is given to the design of ring laser gyros and magnetohydrodynamic rate sensors, redundancy configurations for component failure protection, improvements of hydraulic actuators made on the component level, integrated actuators, problems of software reliability, lightning considerations, and failure detection methods for component and system failures.
Before a program can fail, a software fault must be executed, that execution must alter the data state, and the incorrect data state must propagate to a state that results directly in an incorrect output. This paper describes a tool called PISCES (developed by Reliable Software Technologies Corporation) for predicting the probability that faults in a particular program location will accomplish all three of these steps causing program failure. PISCES is a tool that is used during software verification and validation to predict a program's testability.
Software reliability models require the sequence of interfailure times from the debugging process as input. It was previously illustrated that using data from replicated debugging could greatly improve reliability predictions. However, inexpensive replication of the debugging process requires the existence of a cheap, fast error detector. Laboratory experiments can be designed around a gold version which is used as an oracle or around an n-version error detector. Unfortunately, software developers can not be expected to have an oracle or to bear the expense of n-versions. A generic technique is being investigated for approximating replicated data by using the partially debugged software as a difference detector. It is believed that the failure rate of each fault has significant dependence on the presence or absence of other faults. Thus, in order to discuss a failure rate for a known fault, the presence or absence of each of the other known faults needs to be specified. Also, in simpler models which use shorter input sequences without sacrificing accuracy are of interest. In fact, a possible gain in performance is conjectured. To investigate these propositions, NASA computers running LIC (RTI) versions are used to generate data. This data will be used to label the debugging graph associated with each version. These labeled graphs will be used to test the utility of a surrogate oracle, to analyze the dependent nature of fault failure rates and to explore the feasibility of reliability models which use the data of only the most recent failures.
Life-critical applications warrant a higher level of software reliability than has yet been achieved. Since it is not certain that traditional methods alone can provide the required ultra reliability, new methods should be examined as supplements or replacements. This paper describes a mathematical counterpart to the traditional process of empirical testing. ORA's Penelope verification system is demonstrated as a tool for evaluating the correctness of Ada software. Grady Booch's Ada calendar utility package, obtained through NASA, was specified in the Larch/Ada language. Formal verification in the Penelope environment established that many of the package's subprograms met their specifications. In other subprograms, failed attempts at verification revealed several errors that had escaped detection by testing.
The development of computational icing simulation methods is making the transition form the research to common place use in design and certification efforts. As such, standards of code management, design validation, and documentation must be adjusted to accommodate the increased expectations of the user community with respect to accuracy, reliability, capability, and usability. This paper discusses these concepts with regard to current and future icing simulation code development efforts as implemented by the Icing Branch of the NASA Lewis Research Center in collaboration with the NASA Lewis Engineering Design and Analysis Division. With the application of the techniques outlined in this paper, the LEWICE ice accretion code has become a more stable and reliable software product.
An automated environment with hardware-in-the-loop has been developed by Rocketdyne Huntsville for test of a real-time control system. The target system of application is the man-rated real-time system which controls the Space Shuttle Main Engines (SSME). The primary use of the environment is software verification and validation, but it is also useful for evaluation and analysis of SSME avionics hardware and mathematical engine models. It provides a test bed for the integration of software and hardware. The principles and skills upon which it operates may be applied to other target systems, such as those requiring hardware-in-the-loop simulation and control system development. Potential applications are in problem domains demanding highly reliable software systems requiring testing to formal requirements and verifying successful transition to/from off-nominal system states.