Análisis de las herramientas ORCC y Vivado HLS para la Síntesis de Modelos de Flujo de Datos RVC-CAL

En este Proyecto Fin de Grado se ha realizado un estudio de cómo generar, a partir de modelos de flujo de datos en RVC-CAL (Reconfigurable Video Coding – CAL Actor Language), modelos VHDL (Versatile Hardware Description Language) mediante Vivado HLS (Vivado High Level Synthesis), incluida en las herramientas disponibles en Vivado de Xilinx. Una vez conseguido el modelo VHDL resultante, la intención es que mediante las herramientas de Xilinx se programe en una FPGA (Field Programmable Gate Array) o el dispositivo Zynq también desarrollado por Xilinx. RVC-CAL es un lenguaje de flujo de datos que describe la funcionalidad de bloques funcionales, denominados actores. Las funcionalidades que desarrolla un actor se definen como acciones, las cuales pueden ser diferentes en un mismo actor. Los actores pueden comunicarse entre sí y formar una red de actores o network. Con Vivado HLS podemos obtener un diseño VHDL a partir de un modelo en lenguaje C. Por lo que la generación de modelos en VHDL a partir de otros en RVC-CAL, requiere una fase previa en la que los modelos en RVC-CAL serán compilados para conseguir su equivalente en lenguaje C. El compilador ORCC (Open RVC-CAL Compiler) es la herramienta que nos permite lograr diseños en lenguaje C partiendo de modelos en RVC-CAL. ORCC no crea directamente el código ejecutable, sino que genera un código fuente disponible para ser compilado por otra herramienta, en el caso de este proyecto, el compilador GCC (Gnu C Compiler) de Linux. En resumen en este proyecto nos encontramos con tres puntos de estudio bien diferenciados, los cuales son: 1. Partimos de modelos de flujo de datos en RVC-CAL, los cuales son compilados por ORCC para alcanzar su traducción en lenguaje C. 2. Una vez conseguidos los diseños equivalentes en lenguaje C, son sintetizados en Vivado HLS para conseguir los modelos en VHDL. 3. Los modelos VHDL resultantes serian manipulados por las herramientas de Xilinx para producir el bitstream que sea programado en una FPGA o en el dispositivo Zynq. En el estudio del segundo punto, nos encontramos con una serie de elementos conflictivos que afectan a la síntesis en Vivado HLS de los diseños en lenguaje C generados por ORCC. Estos elementos están relacionados con la manera que se encuentra estructurada la especificación en C generada por ORCC y que Vivado HLS no puede soportar en determinados momentos de la síntesis. De esta manera se ha propuesto una transformación “manual” de los diseños generados por ORCC que afecto lo menos posible a los modelos originales para poder realizar la síntesis con Vivado HLS y crear el fichero VHDL correcto. De esta forma este documento se estructura siguiendo el modelo de un trabajo de investigación. En primer lugar, se exponen las motivaciones y objetivos que apoyan y se esperan lograr en este trabajo. Seguidamente, se pone de manifiesto un análisis del estado del arte de los elementos necesarios para el desarrollo del mismo, proporcionando los conceptos básicos para la correcta comprensión y estudio del documento. Se realiza una descripción de los lenguajes RVC-CAL y VHDL, además de una introducción de las herramientas ORCC y Vivado, analizando las bondades y características principales de ambas. Una vez conocido el comportamiento de ambas herramientas, se describen las soluciones desarrolladas en nuestro estudio de la síntesis de modelos en RVC-CAL, poniéndose de manifiesto los puntos conflictivos anteriormente señalados que Vivado HLS no puede soportar en la síntesis de los diseños en lenguaje C generados por el compilador ORCC. A continuación se presentan las soluciones propuestas a estos errores acontecidos durante la síntesis, con las cuales se pretende alcanzar una especificación en C más óptima para una correcta síntesis en Vivado HLS y alcanzar de esta forma los modelos VHDL adecuados. Por último, como resultado final de este trabajo se extraen un conjunto de conclusiones sobre todos los análisis y desarrollos acontecidos en el mismo. Al mismo tiempo se proponen una serie de líneas futuras de trabajo con las que se podría continuar el estudio y completar la investigación desarrollada en este documento. ABSTRACT. In this Project it has made a study of how to generate, from data flow models in RVC-CAL (Reconfigurable Video Coding - Actor CAL Language), VHDL models (Versatile Hardware Description Language) by Vivado HLS (Vivado High Level Synthesis), included in the tools available in Vivado of Xilinx. Once achieved the resulting VHDL model, the intention is that by the Xilinx tools programmed in FPGA or Zynq device also developed by Xilinx. RVC-CAL is a dataflow language that describes the functionality of functional blocks, called actors. The functionalities developed by an actor are defined as actions, which may be different in the same actor. Actors can communicate with each other and form a network of actors. With Vivado HLS we can get a VHDL design from a model in C. So the generation of models in VHDL from others in RVC-CAL requires a preliminary phase in which the models RVC-CAL will be compiled to get its equivalent in C. The compiler ORCC (Open RVC-CAL Compiler) is the tool that allows us to achieve designs in C language models based on RVC-CAL. ORCC not directly create the executable code but generates an available source code to be compiled by another tool, in the case of this project, the GCC compiler (GNU C Compiler) of Linux. In short, in this project we find three well-defined points of study, which are: 1. We start from data flow models in RVC-CAL, which are compiled by ORCC to achieve its translation in C. 2. Once you realize the equivalent designs in C, they are synthesized in Vivado HLS for VHDL models. 3. The resulting models VHDL would be manipulated by Xilinx tools to produce the bitstream that is programmed into an FPGA or Zynq device. In the study of the second point, we find a number of conflicting elements that affect the synthesis Vivado HLS designs in C generated by ORCC. These elements are related to the way it is structured specification in C generated ORCC and Vivado HLS cannot hold at certain times of the synthesis. Thus it has proposed a "manual" transformation of designs generated by ORCC that affected as little as possible to the original in order to perform the synthesis Vivado HLS and create the correct file VHDL models. Thus this document is structured along the lines of a research. First, the motivations and objectives that support and hope to reach in this work are presented. Then it shows an analysis the state of the art of the elements necessary for its development, providing the basics for a correct understanding and study of the document. A description of the RVC-CAL and VHDL languages is made, in addition an introduction of the ORCC and Vivado tools, analyzing the advantages and main features of both. Once you know the behavior of both tools, the solutions developed in our study of the synthesis of RVC-CAL models, introducing the conflicting points mentioned above are described that Vivado HLS cannot stand in the synthesis of design in C language generated by ORCC compiler. Below the proposed solutions to these errors occurred during synthesis, with which it is intended to achieve optimum C specification for proper synthesis Vivado HLS and thus create the appropriate VHDL models are presented. Finally, as the end result of this work a set of conclusions on all analyzes and developments occurred in the same are removed. At the same time a series of future lines of work which could continue to study and complete the research developed in this document are proposed.

Avaliação do erro estático da ferramenta de uma máquina fresadora com arquitetura paralela.

Os mecanismos amplamente utilizados em aplicações industriais são de tipo serial, porém há algum tempo vem sendo desenvolvidos estudos sobre as vantagens que os mecanismos de arquitetura paralela oferecem em contraposição com os seriais. Rigidez, precisão, altas frequências naturais e velocidade são algumas características que os mecanismos paralelos atribuem a máquinas já consolidadas na indústria, destinadas principalmente nas operações de manipulação (pick and place). Nesse sentido, é relevante o estudo sobre a funcionalidade em outros tipos de operação como a usinagem e, particularmente o fresamento. Para isto, devem-se ainda explorar e desenvolver as capacidades dos mecanismos paralelos em relação à rigidez e à precisão nas operações mencionadas. Foi desenvolvido previamente o projeto e montagem do protótipo de uma máquina fresadora de arquitetura paralela. Também aracterizado pela redundância na atuação para o posicionamento da ferramenta. Com este intuito, pretende-se no trabalho atual, avaliar o erro estático de posicionamento da ferramenta por métodos experimentais, quantificar os deslocamentos, realizar um mapeamento experimental em diversas configurações dos membros. Por outro lado, pretende-se adaptar um modelo numérico simplificado que possa prever as deformações elásticas em diversas configurações, que contemple o efeito de juntas lineares flexíveis e que de alguma forma ajude a identificar as principais fontes de erro. Para tal, foram elaboradas rotinas de programação que através da cinemática inversa e o uso do método dos elementos finitos tentem prever o que de fato acontece nos experimentos. Foi proposta também uma implementação alternativa para o controle do mecanismo através de um software CNC e a conversão de coordenadas cartesianas em coordenadas dos atuadores, isto ajudaria na geração do código G. Finalmente, foram elaboradas algumas trajetórias que tentam avaliar a exatidão e repetitividade do mecanismo além de descrever outras trajetórias livres.

Um método de otimização da relação desempenho/consumo de energia para arquiteturas multi-cores heterogêneas em FPGA

Devido às tendências de crescimento da quantidade de dados processados e a crescente necessidade por computação de alto desempenho, mudanças significativas estão acontecendo no projeto de arquiteturas de computadores. Com isso, tem-se migrado do paradigma sequencial para o paralelo, com centenas ou milhares de núcleos de processamento em um mesmo chip. Dentro desse contexto, o gerenciamento de energia torna-se cada vez mais importante, principalmente em sistemas embarcados, que geralmente são alimentados por baterias. De acordo com a Lei de Moore, o desempenho de um processador dobra a cada 18 meses, porém a capacidade das baterias dobra somente a cada 10 anos. Esta situação provoca uma enorme lacuna, que pode ser amenizada com a utilização de arquiteturas multi-cores heterogêneas. Um desafio fundamental que permanece em aberto para estas arquiteturas é realizar a integração entre desenvolvimento de código embarcado, escalonamento e hardware para gerenciamento de energia. O objetivo geral deste trabalho de doutorado é investigar técnicas para otimização da relação desempenho/consumo de energia em arquiteturas multi-cores heterogêneas single-ISA implementadas em FPGA. Nesse sentido, buscou-se por soluções que obtivessem o melhor desempenho possível a um consumo de energia ótimo. Isto foi feito por meio da combinação de mineração de dados para a análise de softwares baseados em threads aliadas às técnicas tradicionais para gerenciamento de energia, como way-shutdown dinâmico, e uma nova política de escalonamento heterogeneity-aware. Como principais contribuições pode-se citar a combinação de técnicas de gerenciamento de energia em diversos níveis como o nível do hardware, do escalonamento e da compilação; e uma política de escalonamento integrada com uma arquitetura multi-core heterogênea em relação ao tamanho da memória cache L1.

Método para análise da interação fluido-estrutura em travessas do pré-distribuidor de turbinas hidráulicas.

Um dos grandes desafios enfrentados pelos fabricantes de turbinas hidráulicas é prevenir o aparecimento de vibrações induzidas pelo escoamento nas travessas do pré-distribuidor e pás do rotor. Considerando apenas as travessas, e atribuídos a tais vibrações, foram relatados 28 casos de trincas ou ruídos anormais nas últimas décadas, que acarretaram enormes prejuízos associados a reparos, atrasos e perda de geração. O estado da arte na prevenção destes problemas baseia-se na utilização de sofisticados, e caros, programas comerciais de dinâmica dos fluidos computacional para o cálculo transiente do fenômeno. Este trabalho faz uma ampla revisão bibliográfica e levantamento de eventos de trincas ou ruídos ocorridos em travessas nos últimos 50 anos. Propõe, então, um enfoque alternativo, baseado exclusivamente em ferramentas de código aberto. A partir de hipóteses simplificadoras devidamente justificadas, o problema é formulado matematicamente de forma bidimensional, no plano da seção transversal da travessa, levando em conta a interação fluido-estrutura. Nesta estratégia, as equações de Navier-Stokes são resolvidas pelo método dos elementos finitos por meio da biblioteca gratuita oomph-lib. Um código especial em C++ é desenvolvido para o problema de interação fluido-estrutura, no qual o fenômeno de turbulência é levado em consideração por meio de um algoritmo baseado no modelo de Baldwin-Lomax. O método proposto é validado por meio da comparação dos resultados obtidos com referências e medições disponíveis na literatura, que tratam de problemas de barras retangulares suportadas elasticamente. O trabalho finaliza com a aplicação do método a um estudo de caso envolvendo uma travessa particular.

Particle Swarm Optimization Based Reactive Power Dispatch for Power Networks with Distributed Generation

Reactive power is critical to the operation of the power networks on both safety aspects and economic aspects. Unreasonable distribution of the reactive power would severely affect the power quality of the power networks and increases the transmission loss. Currently, the most economical and practical approach to minimizing the real power loss remains using reactive power dispatch method. Reactive power dispatch problem is nonlinear and has both equality constraints and inequality constraints. In this thesis, PSO algorithm and MATPOWER 5.1 toolbox are applied to solve the reactive power dispatch problem. PSO is a global optimization technique that is equipped with excellent searching capability. The biggest advantage of PSO is that the efficiency of PSO is less sensitive to the complexity of the objective function. MATPOWER 5.1 is an open source MATLAB toolbox focusing on solving the power flow problems. The benefit of MATPOWER is that its code can be easily used and modified. The proposed method in this thesis minimizes the real power loss in a practical power system and determines the optimal placement of a new installed DG. IEEE 14 bus system is used to evaluate the performance. Test results show the effectiveness of the proposed method.

Real-time 3D semi-local surface patch extraction using GPGPU

Feature vectors can be anything from simple surface normals to more complex feature descriptors. Feature extraction is important to solve various computer vision problems: e.g. registration, object recognition and scene understanding. Most of these techniques cannot be computed online due to their complexity and the context where they are applied. Therefore, computing these features in real-time for many points in the scene is impossible. In this work, a hardware-based implementation of 3D feature extraction and 3D object recognition is proposed to accelerate these methods and therefore the entire pipeline of RGBD based computer vision systems where such features are typically used. The use of a GPU as a general purpose processor can achieve considerable speed-ups compared with a CPU implementation. In this work, advantageous results are obtained using the GPU to accelerate the computation of a 3D descriptor based on the calculation of 3D semi-local surface patches of partial views. This allows descriptor computation at several points of a scene in real-time. Benefits of the accelerated descriptor have been demonstrated in object recognition tasks. Source code will be made publicly available as contribution to the Open Source Point Cloud Library.

A methodology for evacuation route planning inside buildings using geospatial technology

Evacuation route planning is a fundamental task for building engineering projects. Safety regulations are established so that all occupants are driven on time out of a building to a secure place when faced with an emergency situation. As an example, Spanish building code requires the planning of evacuation routes on large and, usually, public buildings. Engineers often plan these routes on single building projects, repeatedly assigning clusters of rooms to each emergency exit in a trial-and-error process. But problems may arise for a building complex where distribution and use changes make visual analysis cumbersome and sometimes unfeasible. This problem could be solved by using well-known spatial analysis techniques, implemented as a specialized software able to partially emulate engineer reasoning. In this paper we propose and test an easily reproducible methodology that makes use of free and open source software components for solving a case study. We ran a complete test on a building floor at the University of Alicante (Spain). This institution offers a web service (WFS) that allows retrieval of 2D geometries from any building within its campus. We demonstrate how geospatial technologies and computational geometry algorithms can be used for automating the creation and optimization of evacuation routes. In our case study, the engineers’ task is to verify that the load capacity of each emergency exit does not exceed the standards specified by Spain’s current regulations. Using Dijkstra’s algorithm, we obtain the shortest paths from every room to the most appropriate emergency exit. Once these paths are calculated, engineers can run simulations and validate, based on path statistics, different cluster configurations. Techniques and tools applied in this research would be helpful in the design and risk management phases of any complex building project.

The Emergence of Open Hardware Phenomena in Different Technological Environments

Dannie Jost gave an introductory presentation on the emergence on open hardware phenomena, including synthetic biology and other technological environments at the HEPTech Workshop on Open Hardware on June 13 held at the GSI, Darmstadt (Germany). The workshop was organized by CERN and GSI. This event addressed the OSHW phenomenon and its implications for academia and industry with special attention to knowledge and technology transfer issues. Consideration was given to the various aspects of open source hardware development, and how these are dealt with in academia and industry. Presentations from legal experts, academics, practitioners and business provided input for the discussions and exchange of ideas.

Challenges to the Patent System in face of Open Hardware

Dannie Jost gave a presentation outlining some of the challenges to the patent system presented by open source hardware at the "Open Knowledge Festival", under the topic stream treating open design, hardware, manufacturing and making; September 19, 2012; Helsinki, Finland. This topic stream generated considerable discussion, and it serves to educate an audience that is usually very adverse to patents and copyright, and helps the researcher understand the issuing conflicts surrounding emerging technologies, in particular digital technologies, and the maker movement (digitally enabled).

Semi-global techniques in image matching and change detection with applications to civil and environmental engineering.

This work of thesis wants to present a dissertation of the wide range of modern dense matching algorithms, which are spreading in different application and research fields, with a particular attention to the innovative “Semi-Global” matching techniques. The choice of develop a semi-global numerical code was justified by the need of getting insight on the variables and strategies that affect the algorithm performances with the primary objective of maximizing the method accuracy and efficiency, and the results level of completeness. The dissertation will consist in the metrological characterization of the proprietary implementation of the semi-global matching algorithm, evaluating the influence of several matching variables and functions implemented in the process and comparing the accuracy and completeness of different results (digital surface models, disparity maps and 2D displacement fields) obtained using our code and other commercial and open-source matching programs in a wide variety of application fields.

Evaluation of the quality and promotion of open learning materials

In the UK, Open Learning has been used in industrial training for at least the last decade. Trainers and Open Learning practitioners have been concerned about the quality of the products and services being delivered. The argument put forward in this thesis is that there is ambiguity amongst industrialists over the meanings of `Open Learning' and `Quality in Open Learning'. For clarity, a new definition of Open Learning is proposed which challenges the traditional learner-centred approach favoured by educationalists. It introduces the concept that there are benefits afforded to the trainer/employer/teacher as well as to the learner. This enables a focussed view of what quality in Open Learning really means. Having discussed these issues, a new quantitative method of evaluating Open Learning is proposed. This is based upon an assessment of the degree of compliance with which products meet Parts 1 & 2 of the Open Learning Code of Practice. The vehicle for these research studies has been a commercial contract commissioned by the Training Agency for the Engineering Industry Training Board (EITB) to examine the quality of Open Learning products supplied to the engineering industry. A major part of this research has been the application of the evaluation technique to a range of 67 Open Learning products (in eight subject areas). The findings were that good quality products can be found right across the price range - so can average and poor quality ones. The study also shows quite convincingly that there are good quality products to be found at less than 50. Finally the majority (24 out of 34) of the good quality products were text based.

Reproducible computing with Tzar and Docker

The potential for sharing environmental data and models is huge, but can be challenging for experts without specific programming expertise. We built an open-source, cross-platform framework (‘Tzar’) to run models across distributed machines. Tzar is simple to set up and use, allows dynamic parameter generation and enhances reproducibility by accessing versioned data and code. Combining Tzar with Docker helps us lower the entry barrier further by versioning and bundling all required modules and dependencies, together with the database needed to schedule work.

Fifty shades of innovation – from open toward user, and open collaborative forms of innovation – an overview

Starting from the Schumpeterian producer-driven understanding of innovation, followed by user-generated solutions and understanding of collaborative forms of co-creation, scholars investigated the drivers and the nature of interactions underpinning success in various ways. Innovation literature has gone a long way, where open innovation has attracted researchers to investigate problems like compatibilities of external resources, networks of innovation, or open source collaboration. Openness itself has gained various shades in the different strands of literature. In this paper the author provides with an overview and a draft evaluation of the different models of open innovation, illustrated with some empirical findings from various fields drawn from the literature. She points to the relevance of transaction costs affecting viable forms of (open) innovation strategies of firms, and the importance to define the locus of innovation for further analyses of different firm and interaction level formations.

The effects of intellectual property rights on knowledge spillovers and innovation: An empirical investigation

FDI is believed to be a conduit of new technologies between countries. The first chapter of this dissertation studies the advantages of outward FDI for the home country of multinationals conducting research and development abroad. We use patent citations as a proxy for technology spillovers and we bring empirical evidence that supports the hypothesis that a U.S. subsidiary conducting research and development overseas facilitates the flow of knowledge between its host and home countries.^ The second chapter examines the impact of intellectual property rights (IPR) reforms on the technology flows between the U.S. and host countries of U.S. affiliates. We again use patent citations to examine whether the diffusion of new technology between the host countries and the U.S. is accelerated by the reforms. Our results suggest that the reforms favor innovative efforts of domestic firms in the reforming countries rather than U.S. affiliates efforts. In other words, reforms mediate the technology flows from the U.S. to the reforming countries.^ The third chapter deals with another form of IPR, open source (OS) licenses. These differ in the conditions under which licensors and OS contributors are allowed to modify and redistribute the source code. We measure OS project quality by the speed with which programming bugs are fixed and test whether the license chosen by project leaders influences bug resolution rates. In initial regressions, we find a strong correlation between the hazard of bug resolution and the use of highly restrictive licenses. However, license choices are likely to be endogenous. We instrument license choice using (i) the human language in which contributors operate and (ii) the license choice of the project leaders for a previous project. We then find weak evidence that restrictive licenses adversely affect project success.^

Supporting Run-time Monitoring of UML-RT through Customizable Monitoring Configurations in PapyrusRT

Model Driven Engineering uses the principle that code can automatically be generated from software models which would potentially save time and cost of development. By this methodology, a systems structure and behaviour can be expressed in more abstract, high level terms without some of the accidental complexity that the use of a general purpose language can bring. Models are the actual implementation of the system unlike in traditional software development where models are often used for documentation purposes only. However once the code is generated from the model, testing and debugging activities tend to happen on the code level and the model is not updated. We believe that monitoring on the model level could potentially facilitate quality assurance activities as the errors are detected in the early phase of development. In this thesis, we create a Monitoring Configuration for an open source model driven engineering tool called PapyrusRT in Eclipse. We support the run-time monitoring of UML-RT elements with a tracing tool called LTTng. We annotate the model with monitoring information to be used by the code generator for adding tracepoint statements for the corresponding elements. We provide the option of a timing specification to discover latency errors on the model. We validate the results by creating and tracing real time models in PapyrusRT.