No description available.

No description available.

The Configuration Control and Management of Trainer Systems Software has been recognized as a significant problem for many years. Many software tools are being developed for use as aids in this task. The efforts of the Department of Defense in the development of Ada and an Ada Programming Support Environment (APSE) have strongly addressed the problems of configuration control and management. However, these efforts have been targeted for Ada-based software as it is anticipated it will be the primary language of the DOD. The author feels that FORTRAN based software will require maintenance for many years even after the acceptance of Ada, and that the features of the APSE could be applicable to this task. The Ada Language System (ALS), which will soon be available, is the Army's implementation of the Ada Programming Support Environment. ALS has also been accepted by the Navy. This paper discusses the applicability of the ALS to the problem of FORTRAN-based software configuration control and management. A basic demonstration tool will be developed and analyzed. Ada is a registered trademark of the U.S. Government (Ada Joint Program Office).

<p> In this day and age, many companies struggle with the maintenance of legacy scientific software systems written in outdated programming languages. These languages use low-level control structures, algorithmic operations and cumbersome syntax that make the true meaning of an algorithm difficult to understand. To make matters worse, the process of reverse engineering the algorithm to specification often involves a considerable amount of manual work which is error-prone and time-consuming.</p> <p> This thesis explores a completely automated method of reverse engineering. We apply this method to FORTRAN77 linear algebra software. This software is transformed to an extension of FORTRAN77, which we call Fortran-M. This language allows for high-level mathematical constructs such as sums, products and vector and matrix operations. To serve as a proof-of-concept for this method, we have developed a tool which uses a combination of pattern matching on the source code's abstract syntax tree to recognise low-level control structures, and symbolic analysis to determine the meaning of loops. Once a pattern has been recognised, or a loop's invariant found, we apply transformations to the syntax tree, thus creating a Fortran-M equivalent.</p> / Thesis / Master of Applied Science (MASc)

No description available.

Mémoire (M.Eng.)-- Université du Québec ;a Chicoutimi, 1993. / Résumé disponible sur Internet. CaQCU Document électronique également accessible en format PDF. CaQCU

No description available.

Master of Science / Department of Computing and Information Sciences / Mitchell L. Neilsen / People are more inclined to use tablets instead of other computing devices due to their portability and ease of use. A number of desktop applications are now becoming available as tablet applications, with increasing demand in the market. Android is one of the largest and most popular open source platforms that offer developers complete access to framework APIs in order to develop innovative tablet applications. The objective of this project is to develop an Android application for the U.S. Department of Agriculture (USDA) Structural Design Software. The GUI for this software is developed to run on tablet devices powered by Android platform. The main features provided by the User Interface include: • Allowing the input to be saved in ASCII text format and displaying the simulation results in PDF format • Allowing the user to select the type of project or view help contents for the projects • Allowing the user to build the simulation for the selected type of project • Allowing the user to send the simulation results to an e-mail The backend for this software is supposed to replace the old FORTRAN source files with Java source files. FORTRAN to Java translation is performed using the FORTRAN to Java (F2J) translator. F2J is intended to translate old FORTRAN math libraries, but was not completely successful in translating these FORTRAN programs. To accomplish successful translation, some features (such as Common Blocks, IO operations) were removed from the FORTRAN source files before translation. After successful translation, the removed features were added again to the translated Java source files. The simulation results provided by the software are useful to design engineers to develop new structural designs.

International Telemetering Conference Proceedings / October 26-29, 1998 / Town & Country Resort Hotel and Convention Center, San Diego, California / Dryden real-time flight simulators encompass the simulation of pulse code modulation (PCM) telemetry signals. This paper presents a new method whereby the calibration polynomial (from first to sixth order), representing the conversion from counts to engineering units (EU), is numerically inverted in real time. The result is less than onecount error for valid EU inputs. The Newton-Raphson method is used to numerically invert the polynomial. A reverse linear interpolation between the EU limits is used to obtain an initial value for the desired telemetry count. The method presented here is not new. What is new is how classical numerical techniques are optimized to take advantage of modern computer power to perform the desired calculations in real time. This technique makes the method simple to understand and implement. There are no interpolation tables to store in memory as in traditional methods. The NASA F-15 simulation converts and transmits over 1000 parameters at 80 times/sec. This paper presents algorithm development, FORTRAN code, and performance results.

La parallélisation automatique est une des approches visant une plus grande facilité d'utilisation des ordinateurs parallèles. La parallélisation consiste prendre un programme écrit pour une machine séquentielle (qui n'a qu'un processeur) et de l'adapter une machine parallèle. L'intérêt de faire faire cette parallélisation automatiquement par un programme appelé paralléliseur est qu'on pourrait alors réutiliser tout le code déjà écrit en Fortran pour machine séquentielles, après parallélisation, sur des machines parallèles. Nous n'y sommes pas encore, mais on s'en approche. C'est dans ce cadre que se situe mon travail. Une moitié approximativement de ma thèse est consacrée à la réalisation d'un logiciel qui parallélise automatiquement une classe réduite de programmes (les nids de boucles uniformes qui utilisent des translations comme accès aux tableaux de données) en HPF (High Performance Fortran). J'insiste surtout sur la partie génération de code HPF, qui est la partie la plus novatrice de ce programme. Outre la réalisation de Bouclettes, ma contribution au domaine est aussi théorique avec une étude sur un partitionnement des données appelé pavage par des parallélépipèdes et une étude de l'optimisation des calculs d' « expressions de tableaux » dans le langage High Performance Fortran. Le pavage est une technique permettant d'optimiser la taille des tâches qu'on répartit sur les processeurs pour diminuer le temps passé en communications. L'évaluation d'expressions de tableaux est une étape d'optimisation du compilateur parallèle (le programme qui traduit le code parallèle écrit dans un langage de haut niveau comme HPF en code machine directement exécutable par l'ordinateur parallèle).