Massively parallel programming is an increasingly growing field with the recent introduction of general purpose GPU computing. Modern graphics processors from NVIDIA and AMD have massively parallel architectures that can be used for such applications as 3D rendering, financial analysis, physics simulations, and biomedical analysis. These massively parallel systems are exposed to programmers through in- terfaces such as NVIDIAs CUDA, OpenCL, and Microsofts C++ AMP. These frame- works expose functionality using primarily either C or C++. In order to use these massively parallel frameworks, programs being implemented must be run on machines equipped with massively parallel hardware. These requirements limit the flexibility of new massively parallel systems. This paper explores the possibility that massively parallel systems can be exposed through web services in order to facilitate using these architectures from remote systems written in other languages. To explore this possi- bility, an architecture is put forth with requirements and high level design for building a web service that can overcome limitations of existing tools and frameworks. The CUDA Web API is built using Python, PyCUDA, NumPy, JSON, and Django to meet the requirements set forth. Additionaly, a client application, CUDA Cloud, is built and serves as an example web service client. The CUDA Web API’s performance and its functionality is validated using a common matrix multiplication algorithm implemented using different languages and tools. Performance tests show runtime improvements for larger datasets using the CUDA Web API for remote CUDA kernel execution over serial implementations. This paper concludes that existing limitations associated with GPGPU usage can be overcome with the specified architecture.

The use of Conversational Agents (CAs) utilizing Artificial Intelligence Markup Language (AIML) has been studied in a number of disciplines. Previous research has shown a great deal of promise. It has also documented significant limitations in the abilities of these CAs. Many of these limitations are related specifically to the method employed by AIML to resolve ambiguities in the meaning and context of words. While methods exist to detect and correct common errors in spelling and grammar of sentences and queries submitted by a user, one class of input error that is particularly difficult to detect and correct is the malapropism. In this research a malapropism is defined a "verbal blunder in which one word is replaced by another similar in sound but different in meaning" ("malapropism," 2013). This research explored the use of alternative methods of correcting malapropisms in sentences input to AIML CAs using measures of Semantic Distance and tri-gram probabilities. Results of these alternate methods were compared against AIML CAs using only the Symbolic Reductions built into AIML. This research found that the use of the two methodologies studied here did indeed lead to a small, but measurable improvement in the performance of the CA in terms of the appropriateness of its responses as classified by human judges. However, it was also noted that in a large number of cases, the CA simply ignored the existence of a malapropism altogether in formulating its responses. In most of these cases, the interpretation and response to the user's input was of such a general nature that one might question the overall efficacy of the AIML engine. The answer to this question is a matter for further study.

<p>Ова докторска теза описује и анализира приступ развоју Це компајлера за<br />наменске процесоре. Такав компајлер захтева имплементацију посебних<br />техника и алгоритама, претежно специфичних за нерегуларне процесорске<br />архитектуре, да би генерисао ефикасан код, и при том је потребно да<br />испуњава индустријске стандарде по питању робустности, разумљивости<br />кода, могућности одржавања и проширивости. У ту сврху је предложена<br />нова компајлерска инфраструктура над којом је имплементиран компајлер<br />за Cirrus Coyote 32 ДСП. Квалитет генерисаног кода поређен је са<br />квалитетом кода генерисног од стране већ постојећег компајлера за тај<br />процесор. Уједно, одређени елементи организације компајлера су<br />упоређени са популарним компајлерима отвореног кода GCC и LLVM.</p> / <p>Ova doktorska teza opisuje i analizira pristup razvoju Ce kompajlera za<br />namenske procesore. Takav kompajler zahteva implementaciju posebnih<br />tehnika i algoritama, pretežno specifičnih za neregularne procesorske<br />arhitekture, da bi generisao efikasan kod, i pri tom je potrebno da<br />ispunjava industrijske standarde po pitanju robustnosti, razumljivosti<br />koda, mogućnosti održavanja i proširivosti. U tu svrhu je predložena<br />nova kompajlerska infrastruktura nad kojom je implementiran kompajler<br />za Cirrus Coyote 32 DSP. Kvalitet generisanog koda poređen je sa<br />kvalitetom koda generisnog od strane već postojećeg kompajlera za taj<br />procesor. Ujedno, određeni elementi organizacije kompajlera su<br />upoređeni sa popularnim kompajlerima otvorenog koda GCC i LLVM.</p> / <p>This PhD thesis describes and analyses an approach to development of C<br />language compiler for embedded processors. That kind of compiler requires<br />implementation of special techniques and algorithms, mostly specific for<br />irregular processor architectures, in order to be able to generate efficient<br />code, whereas still meeting industrial strength standard by beeing robust,<br />understandable, maintainable, and extensible. For this purpose the new<br />compiler insfrastructure is proposed and on top of it a compiler for Cirrus<br />Logic Coyote 32 DSP is built. Quality of the code generated by that compiler<br />is compared with code generated by the previous compiler for the same<br />processor architecture. Some elements of the compiler design are also<br />compared to popular open source compilers GCC and LLVM.</p>

Traditional archival media such as paper, film, photographs, etc. contain a vast storage of knowledge. Much of this knowledge is applicable to current business and scientific problems, and offers solutions; consequently, there is value in extracting this information. While it is possible to manually extract the content, this technique is not feasible for large knowledge repositories due to cost and time. In this thesis, we develop a system that can extract such knowledge automatically from large repositories. A Graphical User Interface that permits users to indicate the location of the knowledge components (indexes) is developed, and software features that permit automatic extraction of indexes from similar documents is presented. The indexes and the documents are stored in a persistentdata store.The system is tested on a University Registrar’s legacy paper-based transcript repository. The study shows that the system provides a good solution for large-scale extraction of knowledge from archived paper and other media.

[pt] A linguagem Java foi definida tendo como foco a portabilidade. O código gerado pela compilação é interpretado por uma máquina virtual, e não diretamente pelo processador destino, como um programa em C. Este código intermediário, também conhecido como bytecode, é a chave da portabilidade de Java. Os Bytecodes Java usam uma pilha para manipular os operandos das instruções. O uso de pilha tem suas vantagens e desvantagens. Dentre as vantagens, podemos citar a simplicidade da implementação do compilador e da máquina virtual. A principal desvantagem é a redução na velocidade de execução dos programas, devido à necessidade de se mover os operandos para a pilha e retirar dela o resultado, gerando um aumento no número de instruções que devem ser processadas. Diversos estudos indicam que máquinas virtuais baseadas em registradores podem ser mais rápidas que as baseadas em pilha. Decidimos criar uma nova especificação de bytecodes, específicos para máquinas virtuais baseadas em registradores. Esperamos com isso obter um aumento no desempenho das aplicações. / [en] The Java language was created with a focus on portability. The code generated by the compiler is interpreted by a virtual machine, and not directly by the target processor, like programs written in C. This intermediate code, also known as bytecode, is the key to Java's portability. The Java Bytecodes use a stack to manipulate the instruction operands. The use of stack has its their pros and cons. Among the advantages, we can cite the simplicity of implementation of the compiler and virtual machine. On the other hand, there is a speed reduction in the program's execution, due to the need to move the operands to and from the stack, and retrieve results from it, increasing the number of instructions that are processed. Much study has been done that indicating that register-based virtual machines can be faster than the ones based on stacks. Based on this, we decided to create a new bytecode specification, proper for a virtual machine based on registers. By doing this, we hope to obtain an increase in an application's performance.

Devido ao grande número de otimizações fornecidas pelos compiladores modernos e à ampla possibilidade de ordenação dessas transformações, uma eficiente Exploração do Espaço de Projeto (DSE) se faz necessária para procurar a melhor sequência de otimização de uma determinada função ou fragmento de código. Como esta exploração é uma tarefa complexa e dispendiosa, apresentamos uma nova abordagem de DSE capaz de reduzir esse tempo de exploração e selecionar sequências de otimização que melhoraram o desempenho dos códigos transformados. Nossa abordagem utiliza um conjunto de funções de referência, para as quais uma representação simbólica do código (DNA) e a melhor sequência de otimização são conhecidas. O DSE de novas funções é baseado em uma abordagem de agrupamento aplicado sobre o código DNA que identifica similaridades entre funções. O agrupamento utiliza três técnicas para a mineração de dados: distância de compressão normalizada, algoritmo de reconstrução de árvores filogenéticas (Neighbor Joining) e identificação de grupos por ambiguidade. As otimizações das funções de referência identificadas como similares formam o espaço que é explorado para encontrar a melhor sequência para a nova função. O DSE pode utilizar o conjunto reduzido de otimizações de duas formas: como o espaço de projeto ou como a configuração inicial do algoritmo. Em ambos os casos, a adoção de uma pré-seleção baseada no agrupamento permite o uso de algoritmos de busca simples e rápidos. Os resultados experimentais revelam que a nova abordagem resulta numa redução significativa no tempo total de exploração, ao mesmo tempo que alcança um desempenho próximo ao obtido através de uma busca mais extensa e dispendiosa baseada em algoritmos genéticos. / Due to the large number of optimizations provided in modern compilers and to compiler optimization specific opportunities, a Design Space Exploration (DSE) is necessary to search for the best sequence of compiler optimizations for a given code fragment (e.g., function). As this exploration is a complex and time consuming task, we present new DSE strategies to reduce the exploration time and still select optimization sequences able to improve the performance of each function. The DSE is based on a clustering approach which groups functions with similarities and then explore the reduced search space provided by the optimizations previously suggested for the functions in each group. The identification of similarities between functions uses a data mining method which is applied to a symbolic representation of the source code. The DSE strategies uses the reduced optimizations set identified by clustering in two ways: as the design space or as the initial configuration of the algorithm. In both ways, the adoption of a pre-selection based on clustering allows the use of simple and fast DSE algorithms. Several experiments for evaluating the effectiveness of the proposed approach address the exploration of compiler optimization sequences. Besides, we investigate the impact of each technique or component employed in the selection process. Experimental results reveal that the use of our new clustering-based DSE approach achieved a significant reduction on the total exploration time of the search space at the same time that obtained performance speedups close to a traditional genetic algorithmbased approach.

Made available in DSpace on 2015-03-05T14:01:49Z (GMT). No. of bitstreams: 0 Previous issue date: 22 / Bolsa para curso e programa de Pós Graduação / O presente trabalho propõe o Ubiquitous Oriented Programming (abreviadamente UOP), um modelo de programação orientado ao desenvolvimento de sistemas ubíquos. UOP utiliza os conceitos de Serviços e Orientação a Objetos, integrando-os com os requisitos de aplicações ubíquas como contexto, sensibilidade ao contexto, adaptação ao contexto, mobilidade de código e concorrência. Uma linguagem de programação (UbiLanguage) implementa os conceitos deste novo modelo de programação, provendo assim suporte ao desenvolvimento de sistemas ubíquos. O ambiente de desenvolvimento é composto por um compilador (UbiCompiler) que traduz código escrito em UbiLanguage para bytecode. A plataforma de execução é composta por uma máquina virtual ubíqua (UbiVM) que suporta a execução deste bytecode. / This work presents the Ubiquitous Oriented Programming (UOP in short), a programming model oriented to development of ubiquitous systems. UOP uses Services concepts and Object Oriented Programming, integrating them with the requirements of ubiquitous applications as context, context awareness, context adaptation, code mobility and concurrent. A programming language (UbiLanguage) implements the concepts of this new programming model, thus providing the development of ubiquitous systems. The development environment is composed by a compiler (UbiCompiler) which translates code written in UbiLanguage to bytecode. The execution platform is composed by a virtual machine (UbiVM) which supports the execution of this bytecode.

This thesis discusses PRIMM which stands for Parallel Remote Interactive Management Model. PRIMM is a framework for object oriented applications that relies on grid computing. It works as an interface between the remote applications and the parallel computing system. The thesis shows the capabilities that could be achieved from PRIMM architecture.

In this modern rather interconnected era, an organization’s top priority is to protect itself from major security breaches occurring frequently within a communicational environment. But, it seems, as if they quite fail in doing so. Every week there are new headlines relating to information being forged, funds being stolen and corrupt usage of credit card and so on. Personal computers are turned into “zombie machines” by hackers to steal confidential and financial information from sources without disclosing hacker’s true identity. These identity thieves rob private data and ruin the very purpose of privacy. The purpose of this project is to identify suspicious user activity by analyzing a log file which then later can help an investigation agency like FBI to track and monitor anonymous user(s) who seek for weaknesses to attack vulnerable parts of a system to have access of it. The project also emphasizes the potential damage that a malicious activity could have on the system. This project uses Hadoop framework to search and store log files for logging activities and then performs a ‘Map Reduce’ programming code to finally compute and analyze the results.

Computers have been assisting in recording, sound synthesis and other fields of music production for quite some time. The actual performance of music continues to be an area in which human players are chosen over computer performers. Musical performance is an area in which personalization is more important than consistency. Human players play with each other, reacting to phrases and ideas created by the players that they are playing with. Computer performers lack the ability to react to the changes in the performance that humans perceive naturally, giving the human players an advantage over the computer performers. This thesis creates a framework for describing unique musical performers that can play along in realtime with human players. FrTime, a reactive programming language, is used to constantly create new musical phrases. Musical phrases are constructed by unique user programmed performers and by chord changes that the framework provides. The reactive language creates multiple musical phrases for each point in time. A simple module which chooses musical phrases to be performed at the time of performance is created.