An Efficient Parallel Algorithm for Multiple Sequence Similarities Calculation Using a Low Complexity Method

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Improvements in the score matrix calculation method using parallel score estimating algorithm

The increasing amount of sequences stored in genomic databases has become unfeasible to the sequential analysis. Then, the parallel computing brought its power to the Bioinformatics through parallel algorithms to align and analyze the sequences, providing improvements mainly in the running time of these algorithms. In many situations, the parallel strategy contributes to reducing the computational complexity of the big problems. This work shows some results obtained by an implementation of a parallel score estimating technique for the score matrix calculation stage, which is the first stage of a progressive multiple sequence alignment. The performance and quality of the parallel score estimating are compared with the results of a dynamic programming approach also implemented in parallel. This comparison shows a significant reduction of running time. Moreover, the quality of the final alignment, using the new strategy, is analyzed and compared with the quality of the approach with dynamic programming.

Multiobjective Genetic Algorithm applied to dengue control

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Parallel Layout Construction Algorithm for Irregular Shape Packing Problems

Cutting and packing problems arise in a variety of industries, including garment, wood and shipbuilding. Irregular shape packing is a special case which admits irregular items and is much more complex due to the geometry of items. In order to ensure that items do not overlap and no item from the layout protrudes from the container, the collision free region concept was adopted. It represents all possible translations for a new item to be inserted into a container with already placed items. To construct a feasible layout, collision free region for each item is determined through a sequence of Boolean operations over polygons. In order to improve the speed of the algorithm, a parallel version of the layout construction was proposed and it was applied to a simulated annealing algorithm used to solve bin packing problems. Tests were performed in order to determine the speed improvement of the parallel version over the serial algorithm

Performance evaluation of a parallel algorithm for simultaneous untangling and smoothing of tetrahedral meshes

[EN]A new parallel algorithm for simultaneous untangling and smoothing of tetrahedral meshes is proposed in this paper. We provide a detailed analysis of its performance on shared-memory many-core computer architectures. This performance analysis includes the evaluation of execution time, parallel scalability, load balancing, and parallelism bottlenecks. Additionally, we compare the impact of three previously published graph coloring procedures on the performance of our parallel algorithm. We use six benchmark meshes with a wide range of sizes. Using these experimental data sets, we describe the behavior of the parallel algorithm for different data sizes. We demonstrate that this algorithm is highly scalable when it runs on two different high-performance many-core computers with up to 128 processors...

Analysis of optimal control problems for the incompressible MHD equations and implementation in a finite element multiphysics code

This thesis deals with the study of optimal control problems for the incompressible Magnetohydrodynamics (MHD) equations. Particular attention to these problems arises from several applications in science and engineering, such as fission nuclear reactors with liquid metal coolant and aluminum casting in metallurgy. In such applications it is of great interest to achieve the control on the fluid state variables through the action of the magnetic Lorentz force. In this thesis we investigate a class of boundary optimal control problems, in which the flow is controlled through the boundary conditions of the magnetic field. Due to their complexity, these problems present various challenges in the definition of an adequate solution approach, both from a theoretical and from a computational point of view. In this thesis we propose a new boundary control approach, based on lifting functions of the boundary conditions, which yields both theoretical and numerical advantages. With the introduction of lifting functions, boundary control problems can be formulated as extended distributed problems. We consider a systematic mathematical formulation of these problems in terms of the minimization of a cost functional constrained by the MHD equations. The existence of a solution to the flow equations and to the optimal control problem are shown. The Lagrange multiplier technique is used to derive an optimality system from which candidate solutions for the control problem can be obtained. In order to achieve the numerical solution of this system, a finite element approximation is considered for the discretization together with an appropriate gradient-type algorithm. A finite element object-oriented library has been developed to obtain a parallel and multigrid computational implementation of the optimality system based on a multiphysics approach. Numerical results of two- and three-dimensional computations show that a possible minimum for the control problem can be computed in a robust and accurate manner.

A parallel community detection algorithm

Complex networks analysis is a very popular topic in computer science. Unfortunately this networks, extracted from different contexts, are usually very large and the analysis may be very complicated: computation of metrics on these structures could be very complex. Among all metrics we analyse the extraction of subnetworks called communities: they are groups of nodes that probably play the same role within the whole structure. Communities extraction is an interesting operation in many different fields (biology, economics,...). In this work we present a parallel community detection algorithm that can operate on networks with huge number of nodes and edges. After an introduction to graph theory and high performance computing, we will explain our design strategies and our implementation. Then, we will show some performance evaluation made on a distributed memory architectures i.e. the supercomputer IBM-BlueGene/Q "Fermi" at the CINECA supercomputing center, Italy, and we will comment our results.

Parallel algorithm for solving integer linear programs

Linear programs, or LPs, are often used in optimization problems, such as improving manufacturing efficiency of maximizing the yield from limited resources. The most common method for solving LPs is the Simplex Method, which will yield a solution, if one exists, but over the real numbers. From a purely numerical standpoint, it will be an optimal solution, but quite often we desire an optimal integer solution. A linear program in which the variables are also constrained to be integers is called an integer linear program or ILP. It is the focus of this report to present a parallel algorithm for solving ILPs. We discuss a serial algorithm using a breadth-first branch-and-bound search to check the feasible solution space, and then extend it into a parallel algorithm using a client-server model. In the parallel mode, the search may not be truly breadth-first, depending on the solution time for each node in the solution tree. Our search takes advantage of pruning, often resulting in super-linear improvements in solution time. Finally, we present results from sample ILPs, describe a few modifications to enhance the algorithm and improve solution time, and offer suggestions for future work.

An improved bit parallel exact maximum clique algorithm

This paper describes new improvements for BB-MaxClique (San Segundo et al. in Comput Oper Resour 38(2):571–581, 2011 ), a leading maximum clique algorithm which uses bit strings to efficiently compute basic operations during search by bit masking. Improvements include a recently described recoloring strategy in Tomita et al. (Proceedings of the 4th International Workshop on Algorithms and Computation. Lecture Notes in Computer Science, vol 5942. Springer, Berlin, pp 191–203, 2010 ), which is now integrated in the bit string framework, as well as different optimization strategies for fast bit scanning. Reported results over DIMACS and random graphs show that the new variants improve over previous BB-MaxClique for a vast majority of cases. It is also established that recoloring is mainly useful for graphs with high densities.

SLBN: A Scalable Max-min Fair Algorithm for Rate-Based Explicit Congestion Control

The growth of the Internet has increased the need for scalable congestion control mechanisms in high speed networks. In this context, we propose a rate-based explicit congestion control mechanism with which the sources are provided with the rate at which they can transmit. These rates are computed with a distributed max-min fair algorithm, SLBN. The novelty of SLBN is that it combines two interesting features not simultaneously present in existing proposals: scalability and fast convergence to the max-min fair rates, even under high session churn. SLBN is scalable because routers only maintain a constant amount of state information (only three integer variables per link) and only incur a constant amount of computation per protocol packet, independently of the number of sessions that cross the router. Additionally, SLBN does not require processing any data packet, and it converges independently of sessions' RTT. Finally, by design, the protocol is conservative when assigning rates, even in the presence of high churn, which helps preventing link overshoots in transient periods. We claim that, with all these features, our mechanism is a good candidate to be used in real deployments.

Advanced Control Strategies for a 6 DoF Hydraulic Parallel Robot Based on the Dynamic Model

Nowadays robots have made their way into real applications that were prohibitive and unthinkable thirty years ago. This is mainly due to the increase in power computations and the evolution in the theoretical field of robotics and control. Even though there is plenty of information in the current literature on this topics, it is not easy to find clear concepts of how to proceed in order to design and implement a controller for a robot. In general, the design of a controller requires of a complete understanding and knowledge of the system to be controlled. Therefore, for advanced control techniques the systems must be first identified. Once again this particular objective is cumbersome and is never straight forward requiring of great expertise and some criteria must be adopted. On the other hand, the particular problem of designing a controller is even more complex when dealing with Parallel Manipulators (PM), since their closed-loop structures give rise to a highly nonlinear system. Under this basis the current work is developed, which intends to resume and gather all the concepts and experiences involve for the control of an Hydraulic Parallel Manipulator. The main objective of this thesis is to provide a guide remarking all the steps involve in the designing of advanced control technique for PMs. The analysis of the PM under study is minced up to the core of the mechanism: the hydraulic actuators. The actuators are modeled and experimental identified. Additionally, some consideration regarding traditional PID controllers are presented and an adaptive controller is finally implemented. From a macro perspective the kinematic and dynamic model of the PM are presented. Based on the model of the system and extending the adaptive controller of the actuator, a control strategy for the PM is developed and its performance is analyzed with simulation.

Multiphase Design and Control Techniques Applied to a Forward Micro-Inverter

En la última década la potencia instalada de energía solar fotovoltaica ha crecido una media de un 49% anual y se espera que alcance el 16%del consumo energético mundial en el año 2050. La mayor parte de estas instalaciones se corresponden con sistemas conectados a la red eléctrica y un amplio porcentaje de ellas son instalaciones domésticas o en edificios. En el mercado ya existen diferentes arquitecturas para este tipo de instalaciones, entre las que se encuentras los módulos AC. Un módulo AC consiste en un inversor, también conocido como micro-inversor, que se monta en la parte trasera de un panel o módulo fotovoltaico. Esta tecnología ofrece modularidad, redundancia y la extracción de la máxima potencia de cada panel solar de la instalación. Además, la expansión de esta tecnología posibilitará una reducción de costes asociados a las economías de escala y a la posibilidad de que el propio usuario pueda componer su propio sistema. Sin embargo, el micro-inversor debe ser capaz de proporcionar una ganancia de tensión adecuada para conectar el panel solar directamente a la red, mientras mantiene un rendimiento aceptable en un amplio rango de potencias. Asimismo, los estándares de conexión a red deber ser satisfechos y el tamaño y el tiempo de vida del micro-inversor son factores que han de tenerse siempre en cuenta. En esta tesis se propone un micro-inversor derivado de la topología “forward” controlado en el límite entre los modos de conducción continuo y discontinuo (BCM por sus siglas en inglés). El transformador de la topología propuesta mantiene la misma estructura que en el convertidor “forward” clásico y la utilización de interruptores bidireccionales en el secundario permite la conexión directa del inversor a la red. Asimismo el método de control elegido permite obtener factor de potencia cercano a la unidad con una implementación sencilla. En la tesis se presenta el principio de funcionamiento y los principales aspectos del diseño del micro-inversor propuesto. Con la idea de mantener una solución sencilla y de bajo coste, se ha seleccionado un controlador analógico que está originalmente pensado para controlar un corrector del factor de potencia en el mismo modo de conducción que el micro-inversor “forward”. La tesis presenta las principales modificaciones necesarias, con especial atención a la detección del cruce por cero de la corriente (ZCD por sus siglas en inglés) y la compatibilidad del controlador con la inclusión de un algoritmo de búsqueda del punto de máxima potencia (MPPT por sus siglas en inglés). Los resultados experimentales muestran las limitaciones de la implementación elegida e identifican al transformador como el principal contribuyente a las pérdidas del micro-inversor. El principal objetivo de esta tesis es contribuir a la aplicación de técnicas de control y diseño de sistemas multifase en micro-inversores fotovoltaicos. En esta tesis se van a considerar dos configuraciones multifase diferentes aplicadas al micro-inversor “forward” propuesto. La primera consiste en una variación con conexión paralelo-serie que permite la utilización de transformadores con una relación de vueltas baja, y por tanto bien acoplados, para conseguir una ganancia de tensión adecuada con un mejor rendimiento. Esta configuración emplea el mismo control BCM cuando la potencia extraída del panel solar es máxima. Este método de control implica que la frecuencia de conmutación se incrementa considerablemente cuando la potencia decrece, lo que compromete el rendimiento. Por lo tanto y con la intención de mantener unos bueno niveles de rendimiento ponderado, el micro-inversor funciona en modo de conducción discontinuo (DCM, por sus siglas en inglés) cuando la potencia extraía del panel solar es menor que la máxima. La segunda configuración multifase considerada en esta tesis es la aplicación de la técnica de paralelo con entrelazado. Además se han considerado dos técnicas diferentes para decidir el número de fases activas: dependiendo de la potencia continua extraída del panel solar y dependiendo de la potencia instantánea demandada por el micro-inversor. La aplicación de estas técnicas es interesante en los sistemas fotovoltaicos conectados a la red eléctrica por la posibilidad que brindan de obtener un rendimiento prácticamente plano en un amplio rango de potencia. Las configuraciones con entrelazado se controlan en DCM para evitar la necesidad de un control de corriente, lo que es importante cuando el número de fases es alto. Los núcleos adecuados para todas las configuraciones multifase consideradas se seleccionan usando el producto de áreas. Una vez seleccionados los núcleos se ha realizado un diseño detallado de cada uno de los transformadores. Con la información obtenida de los diseños y los resultados de simulación, se puede analizar el impacto que el número de transformadores utilizados tiene en el tamaño y el rendimiento de las distintas configuraciones. Los resultados de este análisis, presentado en esta tesis, se utilizan posteriormente para comparar las distintas configuraciones. Muchas otras topologías se han presentado en la literatura para abordar los diferentes aspectos a considerar en los micro-inversores, que han sido presentados anteriormente. La mayoría de estas topologías utilizan un transformador de alta frecuencia para solventar el salto de tensión y evitar problemas de seguridad y de puesta a tierra. En cualquier caso, es interesante evaluar si topologías sin aislamiento galvánico son aptas para su utilización como micro-inversores. En esta tesis se presenta una revisión de inversores con capacidad de elevar tensión, que se comparan bajo las mismas especificaciones. El objetivo es proporcionar la información necesaria para valorar si estas topologías son aplicables en los módulos AC. Las principales contribuciones de esta tesis son: • La aplicación del control BCM a un convertidor “forward” para obtener un micro-inversor de una etapa sencillo y de bajo coste. • La modificación de dicho micro-inversor con conexión paralelo-series de transformadores que permite reducir la corriente de los semiconductores y una ganancia de tensión adecuada con transformadores altamente acoplados. • La aplicación de técnicas de entrelazado y decisión de apagado de fases en la puesta en paralelo del micro-inversor “forward”. • El análisis y la comparación del efecto en el tamaño y el rendimiento del incremento del número de transformadores en las diferentes configuraciones multifase. • La eliminación de las medidas y los lazos de control de corriente en las topologías multifase con la utilización del modo de conducción discontinuo y un algoritmo MPPT sin necesidad de medida de corriente. • La recopilación y comparación bajo las mismas especificaciones de topologías inversoras con capacidad de elevar tensión, que pueden ser adecuadas para la utilización como micro-inversores. Esta tesis está estructurada en seis capítulos. El capítulo 1 presenta el marco en que se desarrolla la tesis así como el alcance de la misma. En el capítulo 2 se recopilan las topologías existentes de micro-invesores con aislamiento y aquellas sin aislamiento cuya implementación en un módulo AC es factible. Asimismo se presenta la comparación entre estas topologías bajo las mismas especificaciones. El capítulo 3 se centra en el micro-inversor “forward” que se propone originalmente en esta tesis. La aplicación de las técnicas multifase se aborda en los capítulos 4 y 5, en los que se presentan los análisis en función del número de transformadores. El capítulo está orientado a la propuesta paralelo-serie mientras que la configuración con entrelazado se analiza en el capítulo 5. Por último, en el capítulo 6 se presentan las contribuciones de esta tesis y los trabajos futuros. ABSTRACT In the last decade the photovoltaic (PV) installed power increased with an average growth of 49% per year and it is expected to cover the 16% of the global electricity consumption by 2050. Most of the installed PV power corresponds to grid-connected systems, with a significant percentage of residential installations. In these PV systems, the inverter is essential since it is the responsible of transferring into the grid the extracted power from the PV modules. Several architectures have been proposed for grid-connected residential PV systems, including the AC-module technology. An AC-module consists of an inverter, also known as micro-inverter, which is attached to a PV module. The AC-module technology offers modularity, redundancy and individual MPPT of each module. In addition, the expansion of this technology will enable the possibility of economies of scale of mass market and “plug and play” for the user, thus reducing the overall cost of the installation. However, the micro-inverter must be able to provide the required voltage boost to interface a low voltage PV module to the grid while keeping an acceptable efficiency in a wide power range. Furthermore, the quality standards must be satisfied and size and lifetime of the solutions must be always considered. In this thesis a single-stage forward micro-inverter with boundary mode operation is proposed to address the micro-inverter requirements. The transformer in the proposed topology remains as in the classic forward converter and bidirectional switches in the secondary side allows direct connection to the grid. In addition the selected control strategy allows high power factor current with a simple implementation. The operation of the topology is presented and the main design issues are introduced. With the intention to propose a simple and low-cost solution, an analog controller for a PFC operated in boundary mode is utilized. The main necessary modifications are discussed, with the focus on the zero current detection (ZCD) and the compatibility of the controller with a MPPT algorithm. The experimental results show the limitations of the selected analog controller implementation and the transformer is identified as a main losses contributor. The main objective of this thesis is to contribute in the application of control and design multiphase techniques to the PV micro-inverters. Two different multiphase configurations have been applied to the forward micro-inverter proposed in this thesis. The first one consists of a parallel-series connected variation which enables the use of low turns ratio, i.e. well coupled, transformers to achieve a proper voltage boost with an improved performance. This multiphase configuration implements BCM control at maximum load however. With this control method the switching frequency increases significantly for light load operation, thus jeopardizing the efficiency. Therefore, in order to keep acceptable weighted efficiency levels, DCM operation is selected for low power conditions. The second multiphase variation considered in this thesis is the interleaved configuration with two different phase shedding techniques: depending on the DC power extracted from the PV panel, and depending on the demanded instantaneous power. The application of interleaving techniques is interesting in PV grid-connected inverters for the possibility of flat efficiency behavior in a wide power range. The interleaved variations of the proposed forward micro-inverter are operated in DCM to avoid the current loop, which is important when the number of phases is large. The adequate transformer cores for all the multiphase configurations are selected according to the area product parameter and a detailed design of each required transformer is developed. With this information and simulation results, the impact in size and efficiency of the number of transformer used can be assessed. The considered multiphase topologies are compared in this thesis according to the results of the introduced analysis. Several other topological solutions have been proposed to solve the mentioned concerns in AC-module application. The most of these solutions use a high frequency transformer to boost the voltage and avoid grounding and safety issues. However, it is of interest to assess if the non-isolated topologies are suitable for AC-module application. In this thesis a review of transformerless step-up inverters is presented. The compiled topologies are compared using a set benchmark to provide the necessary information to assess whether non-isolated topologies are suitable for AC-module application. The main contributions of this thesis are: • The application of the boundary mode control with constant off-time to a forward converter, to obtain a simple and low-cost single-stage forward micro-inverter. • A modification of the forward micro-inverter with primary-parallel secondary-series connected transformers to reduce the current stress and improve the voltage gain with highly coupled transformers. •The application of the interleaved configuration with different phase shedding strategies to the proposed forward micro-inverter. • An analysis and comparison of the influence in size and efficiency of increasing the number of transformers in the parallel-series and interleaved multiphase configurations. • Elimination of the current loop and current measurements in the multiphase topologies by adopting DCM operation and a current sensorless MPPT. • A compilation and comparison with the same specifications of suitable non-isolated step-up inverters. This thesis is organized in six chapters. In Chapter 1 the background of single-phase PV-connected systems is discussed and the scope of the thesis is defined. Chapter 2 compiles the existing solutions for isolated micro-inverters and transformerless step-up inverters suitable for AC-module application. In addition, the most convenient non-isolated inverters are compared using a defined benchmark. Chapter 3 focuses on the originally proposed single-stage forward micro-inverter. The application of multiphase techniques is addressed in Chapter 4 and Chapter 5, and the impact in different parameters of increasing the number of phases is analyzed. In Chapter 4 an original primary-parallel secondary-series variation of the forward micro-inverter is presented, while Chapter 5 focuses on the application of the interleaved configuration. Finally, Chapter 6 discusses the contributions of the thesis and the future work.

ParAlign: a parallel sequence alignment algorithm for rapid and sensitive database searches

There is a need for faster and more sensitive algorithms for sequence similarity searching in view of the rapidly increasing amounts of genomic sequence data available. Parallel processing capabilities in the form of the single instruction, multiple data (SIMD) technology are now available in common microprocessors and enable a single microprocessor to perform many operations in parallel. The ParAlign algorithm has been specifically designed to take advantage of this technology. The new algorithm initially exploits parallelism to perform a very rapid computation of the exact optimal ungapped alignment score for all diagonals in the alignment matrix. Then, a novel heuristic is employed to compute an approximate score of a gapped alignment by combining the scores of several diagonals. This approximate score is used to select the most interesting database sequences for a subsequent Smith–Waterman alignment, which is also parallelised. The resulting method represents a substantial improvement compared to existing heuristics. The sensitivity and specificity of ParAlign was found to be as good as Smith–Waterman implementations when the same method for computing the statistical significance of the matches was used. In terms of speed, only the significantly less sensitive NCBI BLAST 2 program was found to outperform the new approach. Online searches are available at http://dna.uio.no/search/

An hybrid parallel algorithm for solving tridiagonal linear systems versus the Wang’s method in a Cray T3D BSP computer

In this paper we describe an hybrid algorithm for an even number of processors based on an algorithm for two processors and the Overlapping Partition Method for tridiagonal systems. Moreover, we compare this hybrid method with the Partition Wang’s method in a BSP computer. Finally, we compare the theoretical computation cost of both methods for a Cray T3D computer, using the cost model that BSP model provides.

An algorithm for solving linear recurrence systems on parallel and pipelined machines /

"UILU-ENG 78 1745."