Some techniques for automated, resource-aware distributed and mobile computing in a multi-paradigm programming system

Distributed parallel execution systems speed up applications by splitting tasks into processes whose execution is assigned to different receiving nodes in a high-bandwidth network. On the distributing side, a fundamental problem is grouping and scheduling such tasks such that each one involves sufñcient computational cost when compared to the task creation and communication costs and other such practical overheads. On the receiving side, an important issue is to have some assurance of the correctness and characteristics of the code received and also of the kind of load the particular task is going to pose, which can be specified by means of certificates. In this paper we present in a tutorial way a number of general solutions to these problems, and illustrate them through their implementation in the Ciao multi-paradigm language and program development environment. This system includes facilities for parallel and distributed execution, an assertion language for specifying complex programs properties (including safety and resource-related properties), and compile-time and run-time tools for performing automated parallelization and resource control, as well as certification of programs with resource consumption assurances and efñcient checking of such certificates.

Analysis of service-oriented computing systems

La computación basada en servicios (Service-Oriented Computing, SOC) se estableció como un paradigma ampliamente aceptado para el desarollo de sistemas de software flexibles, distribuidos y adaptables, donde las composiciones de los servicios realizan las tareas más complejas o de nivel más alto, frecuentemente tareas inter-organizativas usando los servicios atómicos u otras composiciones de servicios. En tales sistemas, las propriedades de la calidad de servicio (Quality of Service, QoS), como la rapídez de procesamiento, coste, disponibilidad o seguridad, son críticas para la usabilidad de los servicios o sus composiciones en cualquier aplicación concreta. El análisis de estas propriedades se puede realizarse de una forma más precisa y rica en información si se utilizan las técnicas de análisis de programas, como el análisis de complejidad o de compartición de datos, que son capables de analizar simultáneamente tanto las estructuras de control como las de datos, dependencias y operaciones en una composición. El análisis de coste computacional para la composicion de servicios puede ayudar a una monitorización predictiva así como a una adaptación proactiva a través de una inferencia automática de coste computacional, usando los limites altos y bajos como funciones del valor o del tamaño de los mensajes de entrada. Tales funciones de coste se pueden usar para adaptación en la forma de selección de los candidatos entre los servicios que minimizan el coste total de la composición, basado en los datos reales que se pasan al servicio. Las funciones de coste también pueden ser combinadas con los parámetros extraídos empíricamente desde la infraestructura, para producir las funciones de los límites de QoS sobre los datos de entrada, cuales se pueden usar para previsar, en el momento de invocación, las violaciones de los compromisos al nivel de servicios (Service Level Agreements, SLA) potenciales or inminentes. En las composiciones críticas, una previsión continua de QoS bastante eficaz y precisa se puede basar en el modelado con restricciones de QoS desde la estructura de la composition, datos empiricos en tiempo de ejecución y (cuando estén disponibles) los resultados del análisis de complejidad. Este enfoque se puede aplicar a las orquestaciones de servicios con un control centralizado del flujo, así como a las coreografías con participantes multiples, siguiendo unas interacciones complejas que modifican su estado. El análisis del compartición de datos puede servir de apoyo para acciones de adaptación, como la paralelización, fragmentación y selección de los componentes, las cuales son basadas en dependencias funcionales y en el contenido de información en los mensajes, datos internos y las actividades de la composición, cuando se usan construcciones de control complejas, como bucles, bifurcaciones y flujos anidados. Tanto las dependencias funcionales como el contenido de información (descrito a través de algunos atributos definidos por el usuario) se pueden expresar usando una representación basada en la lógica de primer orden (claúsulas de Horn), y los resultados del análisis se pueden interpretar como modelos conceptuales basados en retículos. ABSTRACT Service-Oriented Computing (SOC) is a widely accepted paradigm for development of flexible, distributed and adaptable software systems, in which service compositions perform more complex, higher-level, often cross-organizational tasks using atomic services or other service compositions. In such systems, Quality of Service (QoS) properties, such as the performance, cost, availability or security, are critical for the usability of services and their compositions in concrete applications. Analysis of these properties can become more precise and richer in information, if it employs program analysis techniques, such as the complexity and sharing analyses, which are able to simultaneously take into account both the control and the data structures, dependencies, and operations in a composition. Computation cost analysis for service composition can support predictive monitoring and proactive adaptation by automatically inferring computation cost using the upper and lower bound functions of value or size of input messages. These cost functions can be used for adaptation by selecting service candidates that minimize total cost of the composition, based on the actual data that is passed to them. The cost functions can also be combined with the empirically collected infrastructural parameters to produce QoS bounds functions of input data that can be used to predict potential or imminent Service Level Agreement (SLA) violations at the moment of invocation. In mission-critical applications, an effective and accurate continuous QoS prediction, based on continuations, can be achieved by constraint modeling of composition QoS based on its structure, known data at runtime, and (when available) the results of complexity analysis. This approach can be applied to service orchestrations with centralized flow control, and choreographies with multiple participants with complex stateful interactions. Sharing analysis can support adaptation actions, such as parallelization, fragmentation, and component selection, which are based on functional dependencies and information content of the composition messages, internal data, and activities, in presence of complex control constructs, such as loops, branches, and sub-workflows. Both the functional dependencies and the information content (described using user-defined attributes) can be expressed using a first-order logic (Horn clause) representation, and the analysis results can be interpreted as a lattice-based conceptual models.

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.

An Aspect-Oriented Approach for Supporting Autonomic Reconfiguration of Software Architecture

The increasing complexity of current software systems is encouraging the development of self-managed software architectures, i.e. systems capable of reconfiguring their structure at runtime to fulfil a set of goals. Several approaches have covered different aspects of their development, but some issues remain open, such as the maintainability or the scalability of self-management subsystems. Centralized approaches, like self-adaptive architectures, offer good maintenance properties but do not scale well for large systems. On the contrary, decentralized approaches, like self-organising architectures, offer good scalability but are not maintainable: reconfiguration specifications are spread and often tangled with functional specifications. In order to address these issues, this paper presents an aspect-oriented autonomic reconfiguration approach where: (1) each subsystem is provided with self-management properties so it can evolve itself and the components that it is composed of; (2) self-management concerns are isolated and encapsulated into aspects, thus improving its reuse and maintenance. Povzetek: Predstavljen je pristop s samo-preoblikovanjem programske arhitekture.

Multi Domain-Specific Modeling of the Security Concerns of Service-Oriented Architectures

Service-Oriented Architectures (SOA), and Web Services (WS), the technology generally used to implement them, achieve the integration of heterogeneous technologies, providing interoperability, and yielding the reutilization of pre-existent systems. Model-driven development methodologies provide inherent benefits such as increased productivity, greater reuse, and better maintainability, to name a few. Efforts on achieving model-driven development of SOAs already exist, but there is currently no standard solution that addresses non-functional aspects of these services as well. This paper presents an approach to integrate these non-functional aspects in the development of web services, with an emphasis on security.

Parallelizing irregular and pointer-based computations automatically: perspectives from logic and constraint programming

Irregular computations pose sorne of the most interesting and challenging problems in automatic parallelization. Irregularity appears in certain kinds of numerical problems and is pervasive in symbolic applications. Such computations often use dynamic data structures, which make heavy use of pointers. This complicates all the steps of a parallelizing compiler, from independence detection to task partitioning and placement. Starting in the mid 80s there has been significant progress in the development of parallelizing compilers for logic pro­gramming (and more recently, constraint programming) resulting in quite capable paralle­lizers. The typical applications of these paradigms frequently involve irregular computations, and make heavy use of dynamic data structures with pointers, since logical variables represent in practice a well-behaved form of pointers. This arguably makes the techniques used in these compilers potentially interesting. In this paper, we introduce in a tutoríal way, sorne of the problems faced by parallelizing compilers for logic and constraint programs and provide pointers to sorne of the significant progress made in the area. In particular, this work has resulted in a series of achievements in the areas of inter-procedural pointer aliasing analysis for independence detection, cost models and cost analysis, cactus-stack memory management, techniques for managing speculative and irregular computations through task granularity control and dynamic task allocation such as work-stealing schedulers), etc.

An Efficient Multiple Object Detection and Tracking Framework for Automatic Counting and Video Surveillance Applications

Automatic visual object counting and video surveillance have important applications for home and business environments, such as security and management of access points. However, in order to obtain a satisfactory performance these technologies need professional and expensive hardware, complex installations and setups, and the supervision of qualified workers. In this paper, an efficient visual detection and tracking framework is proposed for the tasks of object counting and surveillance, which meets the requirements of the consumer electronics: off-the-shelf equipment, easy installation and configuration, and unsupervised working conditions. This is accomplished by a novel Bayesian tracking model that can manage multimodal distributions without explicitly computing the association between tracked objects and detections. In addition, it is robust to erroneous, distorted and missing detections. The proposed algorithm is compared with a recent work, also focused on consumer electronics, proving its superior performance.

Gaussian Multiscale Aggregation oriented to Hand Biometric Segmentation in Mobile Devices

New trends in biometrics are oriented to mobile devices in order to increase the overall security in daily actions like bank account access, e-commerce or even document protection within the mobile. However, applying biometrics to mobile devices imply challenging aspects in biometric data acquisition, feature extraction or private data storage. Concretely, this paper attempts to deal with the problem of hand segmentation given a picture of the hand in an unknown background, requiring an accurate result in terms of hand isolation. For the sake of user acceptability, no restrictions are done on background, and therefore, hand images can be taken without any constraint, resulting segmentation in an exigent task. Multiscale aggregation strategies are proposed in order to solve this problem due to their accurate results in unconstrained and complicated scenarios, together with their properties in time performance. This method is evaluated with a public synthetic database with 480000 images considering different backgrounds and illumination environments. The results obtained in terms of accuracy and time performance highlight their capability of being a suitable solution for the problem of hand segmentation in contact-less environments, outperforming competitive methods in literature like Lossy Data Compression image segmentation (LDC).

Estudio de calidad de codificación utilizando los perfiles 10 Bits del estándar H.264

Desde los inicios de la codificación de vídeo digital hasta hoy, tanto la señal de video sin comprimir de entrada al codificador como la señal de salida descomprimida del decodificador, independientemente de su resolución, uso de submuestreo en los planos de diferencia de color, etc. han tenido siempre la característica común de utilizar 8 bits para representar cada una de las muestras. De la misma manera, los estándares de codificación de vídeo imponen trabajar internamente con estos 8 bits de precisión interna al realizar operaciones con las muestras cuando aún no se han transformado al dominio de la frecuencia. Sin embargo, el estándar H.264, en gran auge hoy en día, permite en algunos de sus perfiles orientados al mundo profesional codificar vídeo con más de 8 bits por muestra. Cuando se utilizan estos perfiles, las operaciones efectuadas sobre las muestras todavía sin transformar se realizan con la misma precisión que el número de bits del vídeo de entrada al codificador. Este aumento de precisión interna tiene el potencial de permitir unas predicciones más precisas, reduciendo el residuo a codificar y aumentando la eficiencia de codificación para una tasa binaria dada. El objetivo de este Proyecto Fin de Carrera es estudiar, utilizando las medidas de calidad visual objetiva PSNR (Peak Signal to Noise Ratio, relación señal ruido de pico) y SSIM (Structural Similarity, similaridad estructural), el efecto sobre la eficiencia de codificación y el rendimiento al trabajar con una cadena de codificación/descodificación H.264 de 10 bits en comparación con una cadena tradicional de 8 bits. Para ello se utiliza el codificador de código abierto x264, capaz de codificar video de 8 y 10 bits por muestra utilizando los perfiles High, High 10, High 4:2:2 y High 4:4:4 Predictive del estándar H.264. Debido a la ausencia de herramientas adecuadas para calcular las medidas PSNR y SSIM de vídeo con más de 8 bits por muestra y un tipo de submuestreo de planos de diferencia de color distinto al 4:2:0, como parte de este proyecto se desarrolla también una aplicación de análisis en lenguaje de programación C capaz de calcular dichas medidas a partir de dos archivos de vídeo sin comprimir en formato YUV o Y4M. ABSTRACT Since the beginning of digital video compression, the uncompressed video source used as input stream to the encoder and the uncompressed decoded output stream have both used 8 bits for representing each sample, independent of resolution, chroma subsampling scheme used, etc. In the same way, video coding standards force encoders to work internally with 8 bits of internal precision when working with samples before being transformed to the frequency domain. However, the H.264 standard allows coding video with more than 8 bits per sample in some of its professionally oriented profiles. When using these profiles, all work on samples still in the spatial domain is done with the same precision the input video has. This increase in internal precision has the potential of allowing more precise predictions, reducing the residual to be encoded, and thus increasing coding efficiency for a given bitrate. The goal of this Project is to study, using PSNR (Peak Signal to Noise Ratio) and SSIM (Structural Similarity) objective video quality metrics, the effects on coding efficiency and performance caused by using an H.264 10 bit coding/decoding chain compared to a traditional 8 bit chain. In order to achieve this goal the open source x264 encoder is used, which allows encoding video with 8 and 10 bits per sample using the H.264 High, High 10, High 4:2:2 and High 4:4:4 Predictive profiles. Given that no proper tools exist for computing PSNR and SSIM values of video with more than 8 bits per sample and chroma subsampling schemes other than 4:2:0, an analysis application written in the C programming language is developed as part of this Project. This application is able to compute both metrics from two uncompressed video files in the YUV or Y4M format.

Strategic directions in constraint programming

An abstract is not available.

Towards execution time estimation in abstract machine-based languages

Abstract machines provide a certain separation between platformdependent and platform-independent concerns in compilation. Many of the differences between architectures are encapsulated in the speciflc abstract machine implementation and the bytecode is left largely architecture independent. Taking advantage of this fact, we present a framework for estimating upper and lower bounds on the execution times of logic programs running on a bytecode-based abstract machine. Our approach includes a one-time, programindependent proflling stage which calculates constants or functions bounding the execution time of each abstract machine instruction. Then, a compile-time cost estimation phase, using the instruction timing information, infers expressions giving platform-dependent upper and lower bounds on actual execution time as functions of input data sizes for each program. Working at the abstract machine level makes it possible to take into account low-level issues in new architectures and platforms by just reexecuting the calibration stage instead of having to tailor the analysis for each architecture and platform. Applications of such predicted execution times include debugging/veriflcation of time properties, certiflcation of time properties in mobile code, granularity control in parallel/distributed computing, and resource-oriented specialization.

User-definable resource bounds analysis for logic programs

We present a static analysis that infers both upper and lower bounds on the usage that a logic program makes of a set of user-definable resources. The inferred bounds will in general be functions of input data sizes. A resource in our approach is a quite general, user-defined notion which associates a basic cost function with elementary operations. The analysis then derives the related (upper- and lower-bound) resource usage functions for all predicates in the program. We also present an assertion language which is used to define both such resources and resourcerelated properties that the system can then check based on the results of the analysis. We have performed some preliminary experiments with some concrete resources such as execution steps, bytes sent or received by an application, number of files left open, number of accesses to a datábase, number of calis to a procedure, number of asserts/retracts, etc. Applications of our analysis include resource consumption verification and debugging (including for mobile code), resource control in parallel/distributed computing, and resource-oriented specialization.

Hiord: A type-free higher-order logic programming language with predicate abstraction

A new formalism, called Hiord, for defining type-free higherorder logic programming languages with predicate abstraction is introduced. A model theory, based on partial combinatory algebras, is presented, with respect to which the formalism is shown sound. A programming language built on a subset of Hiord, and its implementation are discussed. A new proposal for defining modules in this framework is considered, along with several examples.

Improving probabilistic flood forecasting through a data assimilation scheme based on genetic programming

Opportunities offered by high performance computing provide a significant degree of promise in the enhancement of the performance of real-time flood forecasting systems. In this paper, a real-time framework for probabilistic flood forecasting through data assimilation is presented. The distributed rainfall-runoff real-time interactive basin simulator (RIBS) model is selected to simulate the hydrological process in the basin. Although the RIBS model is deterministic, it is run in a probabilistic way through the results of calibration developed in a previous work performed by the authors that identifies the probability distribution functions that best characterise the most relevant model parameters. Adaptive techniques improve the result of flood forecasts because the model can be adapted to observations in real time as new information is available. The new adaptive forecast model based on genetic programming as a data assimilation technique is compared with the previously developed flood forecast model based on the calibration results. Both models are probabilistic as they generate an ensemble of hydrographs, taking the different uncertainties inherent in any forecast process into account. The Manzanares River basin was selected as a case study, with the process being computationally intensive as it requires simulation of many replicas of the ensemble in real time.

Smart Spaces and Smart Objects interoperability Architecture (S3OiA)

The presented work aims to contribute towards the standardization and the interoperability off the Future Internet through an open and scalable architecture design. We present S³OiA as a syntactic/semantic Service-Oriented Architecture that allows the integration of any type of object or device, not mattering their nature, on the Internet of Things. Moreover, the architecture makes possible the use of underlying heterogeneous resources as a substrate for the automatic composition of complex applications through a semantic Triple Space paradigm. Created applications are dynamic and adaptive since they are able to evolve depending on the context where they are executed. The validation scenario of this architecture encompasses areas which are prone to involve human beings in order to promote personal autonomy, such as home-care automation environments and Ambient Assisted Living.