A Web Implementation of A Generalized NEP

The Networks of Evolutionary Processors (NEPs) are computing mechanisms directly inspired from the behavior of cell populations more specifically the point mutations in DNA strands. These mechanisms are been used for solving NP-complete problems by means of a parallel computation postulation. This paper describes an implementation of the basic model of NEP using Web technologies and includes the possibility of designing some of the most common variants of it by means the use of the web page design which eases the configuration of a given problem. It is a system intended to be used in a multicore processor in order to benefit from the multi thread use.

Implementing fully homomorphic encryption schemes in FPGA-based systems

LLas nuevas tecnologías orientadas a la nube, el internet de las cosas o las tendencias "as a service" se basan en el almacenamiento y procesamiento de datos en servidores remotos. Para garantizar la seguridad en la comunicación de dichos datos al servidor remoto, y en el manejo de los mismos en dicho servidor, se hace uso de diferentes esquemas criptográficos. Tradicionalmente, dichos sistemas criptográficos se centran en encriptar los datos mientras no sea necesario procesarlos (es decir, durante la comunicación y almacenamiento de los mismos). Sin embargo, una vez es necesario procesar dichos datos encriptados (en el servidor remoto), es necesario desencriptarlos, momento en el cual un intruso en dicho servidor podría a acceder a datos sensibles de usuarios del mismo. Es más, este enfoque tradicional necesita que el servidor sea capaz de desencriptar dichos datos, teniendo que confiar en la integridad de dicho servidor de no comprometer los datos. Como posible solución a estos problemas, surgen los esquemas de encriptación homomórficos completos. Un esquema homomórfico completo no requiere desencriptar los datos para operar con ellos, sino que es capaz de realizar las operaciones sobre los datos encriptados, manteniendo un homomorfismo entre el mensaje cifrado y el mensaje plano. De esta manera, cualquier intruso en el sistema no podría robar más que textos cifrados, siendo imposible un robo de los datos sensibles sin un robo de las claves de cifrado. Sin embargo, los esquemas de encriptación homomórfica son, actualmente, drás-ticamente lentos comparados con otros esquemas de encriptación clásicos. Una op¬eración en el anillo del texto plano puede conllevar numerosas operaciones en el anillo del texto encriptado. Por esta razón, están surgiendo distintos planteamientos sobre como acelerar estos esquemas para un uso práctico. Una de las propuestas para acelerar los esquemas homomórficos consiste en el uso de High-Performance Computing (HPC) usando FPGAs (Field Programmable Gate Arrays). Una FPGA es un dispositivo semiconductor que contiene bloques de lógica cuya interconexión y funcionalidad puede ser reprogramada. Al compilar para FPGAs, se genera un circuito hardware específico para el algorithmo proporcionado, en lugar de hacer uso de instrucciones en una máquina universal, lo que supone una gran ventaja con respecto a CPUs. Las FPGAs tienen, por tanto, claras difrencias con respecto a CPUs: -Arquitectura en pipeline: permite la obtención de outputs sucesivos en tiempo constante -Posibilidad de tener multiples pipes para computación concurrente/paralela. Así, en este proyecto: -Se realizan diferentes implementaciones de esquemas homomórficos en sistemas basados en FPGAs. -Se analizan y estudian las ventajas y desventajas de los esquemas criptográficos en sistemas basados en FPGAs, comparando con proyectos relacionados. -Se comparan las implementaciones con trabajos relacionados New cloud-based technologies, the internet of things or "as a service" trends are based in data storage and processing in a remote server. In order to guarantee a secure communication and handling of data, cryptographic schemes are used. Tradi¬tionally, these cryptographic schemes focus on guaranteeing the security of data while storing and transferring it, not while operating with it. Therefore, once the server has to operate with that encrypted data, it first decrypts it, exposing unencrypted data to intruders in the server. Moreover, the whole traditional scheme is based on the assumption the server is reliable, giving it enough credentials to decipher data to process it. As a possible solution for this issues, fully homomorphic encryption(FHE) schemes is introduced. A fully homomorphic scheme does not require data decryption to operate, but rather operates over the cyphertext ring, keeping an homomorphism between the cyphertext ring and the plaintext ring. As a result, an outsider could only obtain encrypted data, making it impossible to retrieve the actual sensitive data without its associated cypher keys. However, using homomorphic encryption(HE) schemes impacts performance dras-tically, slowing it down. One operation in the plaintext space can lead to several operations in the cyphertext space. Because of this, different approaches address the problem of speeding up these schemes in order to become practical. One of these approaches consists in the use of High-Performance Computing (HPC) using FPGAs (Field Programmable Gate Array). An FPGA is an integrated circuit designed to be configured by a customer or a designer after manufacturing - hence "field-programmable". Compiling into FPGA means generating a circuit (hardware) specific for that algorithm, instead of having an universal machine and generating a set of machine instructions. FPGAs have, thus, clear differences compared to CPUs: - Pipeline architecture, which allows obtaining successive outputs in constant time. -Possibility of having multiple pipes for concurrent/parallel computation. Thereby, In this project: -We present different implementations of FHE schemes in FPGA-based systems. -We analyse and study advantages and drawbacks of the implemented FHE schemes, compared to related work.

Distributed models in P-Systems architectures to reduce computation time

Membrane systems are computational equivalent to Turing machines. However, their distributed and massively parallel nature obtains polynomial solutions opposite to traditional non-polynomial ones. At this point, it is very important to develop dedicated hardware and software implementations exploiting those two membrane systems features. Dealing with distributed implementations of P systems, the bottleneck communication problem has arisen. When the number of membranes grows up, the network gets congested. The purpose of distributed architectures is to reach a compromise between the massively parallel character of the system and the needed evolution step time to transit from one configuration of the system to the next one, solving the bottleneck communication problem. The goal of this paper is twofold. Firstly, to survey in a systematic and uniform way the main results regarding the way membranes can be placed on processors in order to get a software/hardware simulation of P-Systems in a distributed environment. Secondly, we improve some results about the membrane dissolution problem, prove that it is connected, and discuss the possibility of simulating this property in the distributed model. All this yields an improvement in the system parallelism implementation since it gets an increment of the parallelism of the external communication among processors. Proposed ideas improve previous architectures to tackle the communication bottleneck problem, such as reduction of the total time of an evolution step, increase of the number of membranes that could run on a processor and reduction of the number of processors.

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.

GreenDisc: a HW/SW energy optimization framework in globally distributed computation

In recent future, wireless sensor networks (WSNs) will experience a broad high-scale deployment (millions of nodes in the national area) with multiple information sources per node, and with very specific requirements for signal processing. In parallel, the broad range deployment of WSNs facilitates the definition and execution of ambitious studies, with a large input data set and high computational complexity. These computation resources, very often heterogeneous and driven on-demand, can only be satisfied by high-performance Data Centers (DCs). The high economical and environmental impact of the energy consumption in DCs requires aggressive energy optimization policies. These policies have been already detected but not successfully proposed. In this context, this paper shows the following on-going research lines and obtained results. In the field of WSNs: energy optimization in the processing nodes from different abstraction levels, including reconfigurable application specific architectures, efficient customization of the memory hierarchy, energy-aware management of the wireless interface, and design automation for signal processing applications. In the field of DCs: energy-optimal workload assignment policies in heterogeneous DCs, resource management policies with energy consciousness, and efficient cooling mechanisms that will cooperate in the minimization of the electricity bill of the DCs that process the data provided by the WSNs.

GreenDisc: a HW/SW energy optimization framework in globally distributed computation

In recent future, wireless sensor networks ({WSNs}) will experience a broad high-scale deployment (millions of nodes in the national area) with multiple information sources per node, and with very specific requirements for signal processing. In parallel, the broad range deployment of {WSNs} facilitates the definition and execution of ambitious studies, with a large input data set and high computational complexity. These computation resources, very often heterogeneous and driven on-demand, can only be satisfied by high-performance Data Centers ({DCs}). The high economical and environmental impact of the energy consumption in {DCs} requires aggressive energy optimization policies. These policies have been already detected but not successfully proposed. In this context, this paper shows the following on-going research lines and obtained results. In the field of {WSNs}: energy optimization in the processing nodes from different abstraction levels, including reconfigurable application specific architectures, efficient customization of the memory hierarchy, energy-aware management of the wireless interface, and design automation for signal processing applications. In the field of {DCs}: energy-optimal workload assignment policies in heterogeneous {DCs}, resource management policies with energy consciousness, and efficient cooling mechanisms that will cooperate in the minimization of the electricity bill of the DCs that process the data provided by the WSNs.

Parallel Computer Vision Algorithms for Graphics Processing Units

La evolución de los teléfonos móviles inteligentes, dotados de cámaras digitales, está provocando una creciente demanda de aplicaciones cada vez más complejas que necesitan algoritmos de visión artificial en tiempo real; puesto que el tamaño de las señales de vídeo no hace sino aumentar y en cambio el rendimiento de los procesadores de un solo núcleo se ha estancado, los nuevos algoritmos que se diseñen para visión artificial han de ser paralelos para poder ejecutarse en múltiples procesadores y ser computacionalmente escalables. Una de las clases de procesadores más interesantes en la actualidad se encuentra en las tarjetas gráficas (GPU), que son dispositivos que ofrecen un alto grado de paralelismo, un excelente rendimiento numérico y una creciente versatilidad, lo que los hace interesantes para llevar a cabo computación científica. En esta tesis se exploran dos aplicaciones de visión artificial que revisten una gran complejidad computacional y no pueden ser ejecutadas en tiempo real empleando procesadores tradicionales. En cambio, como se demuestra en esta tesis, la paralelización de las distintas subtareas y su implementación sobre una GPU arrojan los resultados deseados de ejecución con tasas de refresco interactivas. Asimismo, se propone una técnica para la evaluación rápida de funciones de complejidad arbitraria especialmente indicada para su uso en una GPU. En primer lugar se estudia la aplicación de técnicas de síntesis de imágenes virtuales a partir de únicamente dos cámaras lejanas y no paralelas—en contraste con la configuración habitual en TV 3D de cámaras cercanas y paralelas—con información de color y profundidad. Empleando filtros de mediana modificados para la elaboración de un mapa de profundidad virtual y proyecciones inversas, se comprueba que estas técnicas son adecuadas para una libre elección del punto de vista. Además, se demuestra que la codificación de la información de profundidad con respecto a un sistema de referencia global es sumamente perjudicial y debería ser evitada. Por otro lado se propone un sistema de detección de objetos móviles basado en técnicas de estimación de densidad con funciones locales. Este tipo de técnicas es muy adecuada para el modelado de escenas complejas con fondos multimodales, pero ha recibido poco uso debido a su gran complejidad computacional. El sistema propuesto, implementado en tiempo real sobre una GPU, incluye propuestas para la estimación dinámica de los anchos de banda de las funciones locales, actualización selectiva del modelo de fondo, actualización de la posición de las muestras de referencia del modelo de primer plano empleando un filtro de partículas multirregión y selección automática de regiones de interés para reducir el coste computacional. Los resultados, evaluados sobre diversas bases de datos y comparados con otros algoritmos del estado del arte, demuestran la gran versatilidad y calidad de la propuesta. Finalmente se propone un método para la aproximación de funciones arbitrarias empleando funciones continuas lineales a tramos, especialmente indicada para su implementación en una GPU mediante el uso de las unidades de filtraje de texturas, normalmente no utilizadas para cómputo numérico. La propuesta incluye un riguroso análisis matemático del error cometido en la aproximación en función del número de muestras empleadas, así como un método para la obtención de una partición cuasióptima del dominio de la función para minimizar el error. ABSTRACT The evolution of smartphones, all equipped with digital cameras, is driving a growing demand for ever more complex applications that need to rely on real-time computer vision algorithms. However, video signals are only increasing in size, whereas the performance of single-core processors has somewhat stagnated in the past few years. Consequently, new computer vision algorithms will need to be parallel to run on multiple processors and be computationally scalable. One of the most promising classes of processors nowadays can be found in graphics processing units (GPU). These are devices offering a high parallelism degree, excellent numerical performance and increasing versatility, which makes them interesting to run scientific computations. In this thesis, we explore two computer vision applications with a high computational complexity that precludes them from running in real time on traditional uniprocessors. However, we show that by parallelizing subtasks and implementing them on a GPU, both applications attain their goals of running at interactive frame rates. In addition, we propose a technique for fast evaluation of arbitrarily complex functions, specially designed for GPU implementation. First, we explore the application of depth-image–based rendering techniques to the unusual configuration of two convergent, wide baseline cameras, in contrast to the usual configuration used in 3D TV, which are narrow baseline, parallel cameras. By using a backward mapping approach with a depth inpainting scheme based on median filters, we show that these techniques are adequate for free viewpoint video applications. In addition, we show that referring depth information to a global reference system is ill-advised and should be avoided. Then, we propose a background subtraction system based on kernel density estimation techniques. These techniques are very adequate for modelling complex scenes featuring multimodal backgrounds, but have not been so popular due to their huge computational and memory complexity. The proposed system, implemented in real time on a GPU, features novel proposals for dynamic kernel bandwidth estimation for the background model, selective update of the background model, update of the position of reference samples of the foreground model using a multi-region particle filter, and automatic selection of regions of interest to reduce computational cost. The results, evaluated on several databases and compared to other state-of-the-art algorithms, demonstrate the high quality and versatility of our proposal. Finally, we propose a general method for the approximation of arbitrarily complex functions using continuous piecewise linear functions, specially formulated for GPU implementation by leveraging their texture filtering units, normally unused for numerical computation. Our proposal features a rigorous mathematical analysis of the approximation error in function of the number of samples, as well as a method to obtain a suboptimal partition of the domain of the function to minimize approximation error.

Design and Optimization of Power Delivery and Distribution Systems Using Evolutionary Computation Techniques

Nowadays computing platforms consist of a very large number of components that require to be supplied with diferent voltage levels and power requirements. Even a very small platform, like a handheld computer, may contain more than twenty diferent loads and voltage regulators. The power delivery designers of these systems are required to provide, in a very short time, the right power architecture that optimizes the performance, meets electrical specifications plus cost and size targets. The appropriate selection of the architecture and converters directly defines the performance of a given solution. Therefore, the designer needs to be able to evaluate a significant number of options in order to know with good certainty whether the selected solutions meet the size, energy eficiency and cost targets. The design dificulties of selecting the right solution arise due to the wide range of power conversion products provided by diferent manufacturers. These products range from discrete components (to build converters) to complete power conversion modules that employ diferent manufacturing technologies. Consequently, in most cases it is not possible to analyze all the alternatives (combinations of power architectures and converters) that can be built. The designer has to select a limited number of converters in order to simplify the analysis. In this thesis, in order to overcome the mentioned dificulties, a new design methodology for power supply systems is proposed. This methodology integrates evolutionary computation techniques in order to make possible analyzing a large number of possibilities. This exhaustive analysis helps the designer to quickly define a set of feasible solutions and select the best trade-off in performance according to each application. The proposed approach consists of two key steps, one for the automatic generation of architectures and other for the optimized selection of components. In this thesis are detailed the implementation of these two steps. The usefulness of the methodology is corroborated by contrasting the results using real problems and experiments designed to test the limits of the algorithms.

Lock-free Parallel Dynamic Programming

We show a method for parallelizing top down dynamic programs in a straightforward way by a careful choice of a lock-free shared hash table implementation and randomization of the order in which the dynamic program computes its subproblems. This generic approach is applied to dynamic programs for knapsack, shortest paths, and RNA structure alignment, as well as to a state-of-the-art solution for minimizing the máximum number of open stacks. Experimental results are provided on three different modern multicore architectures which show that this parallelization is effective and reasonably scalable. In particular, we obtain over 10 times speedup for 32 threads on the open stacks problem.

Parallel execution of Prolog programs: a survey

Since the early days of logic programming, researchers in the field realized the potential for exploitation of parallelism present in the execution of logic programs. Their high-level nature, the presence of nondeterminism, and their referential transparency, among other characteristics, make logic programs interesting candidates for obtaining speedups through parallel execution. At the same time, the fact that the typical applications of logic programming frequently involve irregular computations, make heavy use of dynamic data structures with logical variables, and involve search and speculation, makes the techniques used in the corresponding parallelizing compilers and run-time systems potentially interesting even outside the field. The objective of this article is to provide a comprehensive survey of the issues arising in parallel execution of logic programming languages along with the most relevant approaches explored to date in the field. Focus is mostly given to the challenges emerging from the parallel execution of Prolog programs. The article describes the major techniques used for shared memory implementation of Or-parallelism, And-parallelism, and combinations of the two. We also explore some related issues, such as memory management, compile-time analysis, and execution visualization.

Dendritic-Like Reliable Computation in Artificial Neurons

Dendritic computation is a term that has been in neuro physiological research for a long time [1]. It is still controversial and far for been clarified within the concepts of both computation and neurophysiology [2], [3]. In any case, it hasnot been integrated neither in a formal computational scheme or structure, nor into formulations of artificial neural nets. Our objective here is to formulate a type of distributed computation that resembles dendritic trees, in such a way that it shows the advantages of neural network distributed computation, mostly the reliability that is shown under the existence of holes (scotomas) in the computing net, without ?blind spots?.

Projector Model for Efficient List-Mode Reconstruction in PET Scanners with Parallel Planar Detectors

We have developed a new projector model specifically tailored for fast list-mode tomographic reconstructions in Positron emission tomography (PET) scanners with parallel planar detectors. The model provides an accurate estimation of the probability distribution of coincidence events defined by pairs of scintillating crystals. This distribution is parameterized with 2D elliptical Gaussian functions defined in planes perpendicular to the main axis of the tube of response (TOR). The parameters of these Gaussian functions have been obtained by fitting Monte Carlo simulations that include positron range, acolinearity of gamma rays, as well as detector attenuation and scatter effects. The proposed model has been applied efficiently to list-mode reconstruction algorithms. Evaluation with Monte Carlo simulations over a rotating high resolution PET scanner indicates that this model allows to obtain better recovery to noise ratio in OSEM (ordered-subsets, expectation-maximization) reconstruction, if compared to list-mode reconstruction with symmetric circular Gaussian TOR model, and histogram-based OSEM with precalculated system matrix using Monte Carlo simulated models and symmetries.

Parallel and Distributed Data Management. Introduction

The manipulation and handling of an ever increasing volume of data by current data-intensive applications require novel techniques for e?cient data management. Despite recent advances in every aspect of data management (storage, access, querying, analysis, mining), future applications are expected to scale to even higher degrees, not only in terms of volumes of data handled but also in terms of users and resources, often making use of multiple, pre-existing autonomous, distributed or heterogeneous resources.

A parallel implementation of 3D Zernike moment analysis

Zernike polynomials are a well known set of functions that find many applications in image or pattern characterization because they allow to construct shape descriptors that are invariant against translations, rotations or scale changes. The concepts behind them can be extended to higher dimension spaces, making them also fit to describe volumetric data. They have been less used than their properties might suggest due to their high computational cost. We present a parallel implementation of 3D Zernike moments analysis, written in C with CUDA extensions, which makes it practical to employ Zernike descriptors in interactive applications, yielding a performance of several frames per second in voxel datasets about 2003 in size. In our contribution, we describe the challenges of implementing 3D Zernike analysis in a general-purpose GPU. These include how to deal with numerical inaccuracies, due to the high precision demands of the algorithm, or how to deal with the high volume of input data so that it does not become a bottleneck for the system.

Particle methods parallel implementations by GP-GPU strategies

This paper outlines the problems found in the parallelization of SPH (Smoothed Particle Hydrodynamics) algorithms using Graphics Processing Units. Different results of some parallel GPU implementations in terms of the speed-up and the scalability compared to the CPU sequential codes are shown. The most problematic stage in the GPU-SPH algorithms is the one responsible for locating neighboring particles and building the vectors where this information is stored, since these specific algorithms raise many dificulties for a data-level parallelization. Because of the fact that the neighbor location using linked lists does not show enough data-level parallelism, two new approaches have been pro- posed to minimize bank conflicts in the writing and subsequent reading of the neighbor lists. The first strategy proposes an efficient coordination between CPU-GPU, using GPU algorithms for those stages that allow a straight forward parallelization, and sequential CPU algorithms for those instructions that involve some kind of vector reduction. This coordination provides a relatively orderly reading of the neighbor lists in the interactions stage, achieving a speed-up factor of x47 in this stage. However, since the construction of the neighbor lists is quite expensive, it is achieved an overall speed-up of x41. The second strategy seeks to maximize the use of the GPU in the neighbor's location process by executing a specific vector sorting algorithm that allows some data-level parallelism. Al- though this strategy has succeeded in improving the speed-up on the stage of neighboring location, the global speed-up on the interactions stage falls, due to inefficient reading of the neighbor vectors. Some changes to these strategies are proposed, aimed at maximizing the computational load of the GPU and using the GPU texture-units, in order to reach the maximum speed-up for such codes. Different practical applications have been added to the mentioned GPU codes. First, the classical dam-break problem is studied. Second, the wave impact of the sloshing fluid contained in LNG vessel tanks is also simulated as a practical example of particle methods