GPGPU Computing using CUDA is rapidly gaining ground today. GPGPU has been brought to the masses through the ease of use of CUDA and ubiquity of graphics cards supporting the same. Although CUDA has a low learning curve for programmers familiar with standard programming languages like C, extracting optimum performance from it, through optimizations and hand tuning is not a trivial task. This is because, in case of GPGPU, an optimization strategy rarely affects the functioning in an isolated manner. Many optimizations affect different aspects for better or worse, establishing a tradeoff situation between them, which needs to be carefully handled to achieve good performance. Thus optimizing an application for CUDA is tough and the performance gain might not be commensurate to the coding effort put in. I propose to simplify the process of optimizing CUDA programs using a CUDA Performance Analyzer. The analyzer is based on analytical modeling of CUDA compatible GPUs. The model characterizes the different aspects of GPU compute unified architecture and can make prediction about expected performance of a CUDA program. It would also give an insight into the performance bottlenecks of the CUDA implementation. This would hint towards, what optimizations need to be applied to improve performance. Based on the model, one would also be able to make a prediction about the performance of the application if the optimizations are applied to the CUDA implementation. This enables a CUDA programmer to test out different optimization strategies without putting in a lot of coding effort.

Thesis (Ph.D.)--Mississippi State University. Department of Computer Science and Engineering. / Title from title screen. Includes bibliographical references.

Devido ao crescente número e diversidade de usuários, novas abordagens de teste são necessárias para reduzir a ocorrência de defeitos e garantir uma melhor qualidade em aplicações móveis. As particularidades desse tipo de software exigem que as técnicas de teste tradicionais sejam revisitadas e novas abordagens propostas. A natureza direcionada a eventos e as funcionalidades de aplicações móveis demandam que os testes sejam executados de maneira automatizada. O Teste Baseado em Modelo (TBM) apresenta-se como uma abordagem válida e promissora que oportuniza o uso de um processo definido, bem como de mecanismos e técnicas formais para o teste de aplicações móveis. Esta dissertação de mestrado tem como objetivo investigar a adoção de TBM junto à técnica de modelagem Event Sequence Graph (ESG) no teste de aplicações móveis para a plataforma Android. Inicialmente, o TBM é avaliado com o apoio de ESG e da ferramenta Robotium. Com base nos resultados obtidos e desafios identificados, propõe-se uma abordagem específica que fundamenta o reuso de modelos de teste para (i) reduzir o esforço manual demandado na etapa de concretização de casos de teste e (ii) testar distintas características inerentes ao contexto de mobilidade. Uma ferramenta de apoio foi projetada e desenvolvida para automatizar a abordagem proposta. Um estudo experimental em ambiente industrial é conduzido para avaliar a abordagem e a ferramenta propostas quanto à efetividade na redução do esforço requisitado para a concretização, bem como à capacidade de detecção de defeitos em aplicações móveis desenvolvidas na plataforma Android. / Due to the increasing number and diversity of users, new testing approaches are necessary to reduce the presence of faults and ensure better quality in mobile applications. The particularities of this class of software require that traditional testing techniques are revisited and new approaches proposed. The event oriented nature and functionalities of mobile applications demand tests that can be performed automatically. Model-Based Testing (MBT) is a valid and promising approach that favors the use of a defined process, as well as mechanisms and formal techniques for the testing of mobile applications. This dissertation investigates the odoption of MBT along with the modeling technique Event Sequence Graph (ESG) to test Android applications. Initially, we evaluate TBM supported by ESG and the Robotium tool. Based on the results and challenges identified, we propose a specific approach underlying the reuse of test models to (i) reduce the manual effort to the concretization of test cases and to (ii) test different and inherited characteristics of the mobility context. A supporting tool was designed and implemented to automate the proposed approach. Finaly, we conducted an experimental study in an industrial environment to evaluate the proposed approach and tool regarding the effectiveness in reducing the concretization’s efforts, as well as the fault detection capability in Android mobile applications.

Contexto: É progressivo e significativo o número de empresas que estão distribuindo seus processos de produção de software ao redor do mundo visando ganhos de produtividade, redução de custos e melhorias na qualidade. Por isso, o Desenvolvimento Distribuído de Software (DDS) tem suscitado grande número de pesquisas na área de engenharia de software nos últimos anos e, os engenheiros, em reconhecimento à grande influência desta forma de trabalho, estão em busca de modelos e processos que facilitem produção de software, com equipes geograficamente dispersas. Além dos engenheiros, gerentes e executivos têm enfrentado desafios e dificuldades em diferentes níveis, envolvendo fatores gerenciais, em especial quanto ao monitoramento de projetos. O monitoramento de projetos de software é determinante para que se obtenha êxito em seu desenvolvimento. Quanto ao desenvolvimento de software, uma técnica que conquistou destaque é o Kanban, empregado junto à metodologias ágeis, neste sentido, é possível ressaltar também o uso de Business Process Model and Notation (BPMN) em modelos que presam pelo gerenciamento e monitoramento de projetos. Objetivo: Nesta perspectiva, o objetivo desta dissertação de mestrado é comprovar a seguinte hipótese: “Um projeto distribuído de software pode ter o cronograma de suas atividades monitoradas, por meio de um processo que aplique a técnica do Kanban e a modelagem BPMN”. Para tal, foi implementado um processo que aplica a técnica mencionada, junto à notação, como instrumento para a validação da hipótese. Métodos e Procedimentos: Com o propósito de comprovar a referida hipótese, foi utilizado o método experimental, o qual contou com 82 participantes, que trabalharam distribuidamente. Resultados: Após a realização do experimento, foram sistematizadas as informações; estes revelaram que a hipótese supracitada é válida, por conseguinte o processo proposto, que aplica as técnicas do Kanban e a modelagem BPMN, é efetivo no monitoramento do cronograma das atividades em um projeto distribuído. Espera-se que, com este resultado, a técnica Kanban, pouco explorada no DDS seja vista como vantajosa, por gerentes de projetos distribuídos, deste modo também melhorar o monitoramento das atividades em projetos desta natureza, e ampliar os materiais de pesquisa na área. / Context: It is progressive and significant number of companies that are distributing their software production processes around the world aiming gatasting productivity, reduced costs and improved quality. Therefore, the Development Distributed Software (DDS) has attracted large number of pesquisas in software engineering in recent years, and engineers, in recognition of the great influence of this form of work, they are looking for models and processes that facilitate production software, with geographically dispersed teams mind. In addition to engineers, managers and executives have faced challenges and difficulties, at different levels, involving managerial factors in spe cial as the project monitoring. Monitoring of software projects is crucial in order to obtain success in its development. As for the development of software, a technique that gained prominence is the Kanban inpreached by the agile methodologies, in this sense, it is also possible to emphasize the use of Business Process Model and Notation (BPMN) models that presam for managing and project monitoring. Objective: In this perspective, the objective of this master thesis is to prove the following hypothesis: "A software distributed design can have cronograma of their activities monitored through a process to apply the technique of Kanban and modeling BPMN. " To this end, it implemented a process that applies the aforementioned technique, with the notation as a tool to validate the hypothesis. Methods and Procedures: In order to prove that mortgage is, we used the experimental method, which was attended by 82 participants, who tra-balharam dis-tribuidamente. Results: After the experiment, were systematized intraining; These revealed that the above hypothesis is valid, therefore the proposed process, which applies the techniques of Kanban and BPMN modeling is efftive in monitoring the schedule of activities in a distributed project. Espear that, with this result, Kanban technique, little explored in DDS is seen as advantageous for project managers distributed thus also me lhorar monitoring activities in projects of this nature, and expand the materials research in the area.

Despite the inherent limitation of accessing information serially, voice applications are increasingly growing in popularity as computing technologies advance. This is a positive development, because voice communication offers a number of benefits over other forms of communication. For example, voice may be better for delivering services to users whose eyes and hands may be engaged in other activities (e.g. driving) or to semi-literate or illiterate users. This thesis proposes a knowledge based architecture for building voice applications to help reduce the limitations of serial access to information. The proposed architecture, called OVR (Ontologies, VoiceXML and Reasoners), uses a rich backend that represents knowledge via ontologies and utilises reasoning engines to reason with it, in order to generate intelligent behaviour. Ontologies were chosen over other knowledge representation formalisms because of their expressivity and executable format, and because current trends suggest a general shift towards the use of ontologies in many systems used for information storing and sharing. For the frontend, this architecture uses VoiceXML, the emerging, and de facto standard for voice automated applications. A functional prototype was built for an initial validation of the architecture. The system is a simple voice application to help locate information about service providers that offer HIV (Human Immunodeficiency Virus) testing. We called this implementation HTLS (HIV Testing Locator System). The functional prototype was implemented using a number of technologies. OWL API, a Java interface designed to facilitate manipulation of ontologies authored in OWL was used to build a customised query interface for HTLS. Pellet reasoner was used for supporting queries to the knowledge base and Drools (JBoss rule engine) was used for processing dialog rules. VXI was used as the VoiceXML browser and an experimental softswitch called iLanga as the bridge to the telephony system. (At the heart of iLanga is Asterisk, a well known PBX-in-a-box.) HTLS behaved properly under system testing, providing the sought initial validation of OVR. / LaTeX with hyperref package

Although there are several Software Development kits and Application Programming Interfaces for client-side location-based services development, they mostly involve the creation of self-referencing location-based services. Self-referencing location-based services include services such as geocoding, reverse geocoding, route management and navigation which focus on satisfying the location-based requirements of a single mobile device. There is a lack of open-source Software Development Kits for the development of client-side location-based services that are cross-referencing. Cross-referencing location-based services are designed for the sharing of location information amongst different entities on a given network. This project was undertaken to assemble, through incremental prototyping, a client-side Java Micro Edition location-based services Software Development Kit and a Mobicents location server to aid mobile network operators and developers alike in the quick creation of the transport and privacy protection of cross-referencing location-based applications on Session Initiation Protocol bearer networks. The privacy of the location information is protected using geolocation policies. Developers do not need to have an understanding of Session Initiation Protocol event signaling specifications or of the XML Configuration Access Protocol to use the tools that we put together. The developed tools are later consolidated using two sample applications, the friend-finder and child-tracker services. Developer guidelines are also provided, to aid in using the provided tools.

Personal Software Process (PSP) Scriber is a Web-based software engineering tool designed to implement an automatic system for performing PSP. The basis of this strategy is a set of tools to facilitate collection and analysis of development data. By analyzing the collected data, the developer can make informed, accurate decisions about their individual development effort.

The project is a web e-mail application to provide a web page interface for all CSCI faculty, staff and students to handle their e-mails. The application is written by JSP, Java Servlets, JavaScript and custom JSP tag libraries. Regular e-mail capabilities have been enhanced by the feature of allowing users to store and manage messages by day (store to daily folders, view in daily folders, append notes for that day).

The purpose of this project is to design, build and implement a web application system for the Olympiad delegation registration. All the pages and user registration information will be stored in a PostgreSQL database and retrieved by JAVA Servlet and JDBC (JAVA Database Connectivity). The main purpose of this project is to provide an easy-to-register and web-base communication evironment for the National Olympic Committes (NOC) and the Organizing Committee for the Olympic Games (OCOG).

This project "Designing Secure, Java Based Online Registration Systems to Meet Peak Load Performance Targets" is a simulation of a Web-based exposition management system plus a performance testing procedure to examine this web application.