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.

Generador de interfaces de usuario como periódicos gráficos interactivos

El proyecto se encuadra en el ámbito de la generación automática de interfaces de usuario, orientados a entornos web y con contenido multimedia, por otro lado dicha generación de interfaces, encaja en la disciplina de simulación del periodismo de datos, concretamente en el área de presentación y difusión. Sobre la generación automática de interfaces de usuario, en este proyecto se considera un tipo de interfaz de usuario a modo de periódico virtual, que permita al usuario visualizar de forma clara la información de diferentes noticias y permita la interacción entre los diferentes elementos de dicha presentación. Dicho interfaz de usuario se diseña usando la metáfora periodística, es decir, con un esquema de organización similar al del periódico clásico, para así facilitar la comprensión de las noticias, gracias a un interfaz sencillo y claro. OBJETIVOS El objetivo principal de este proyecto consiste en dar solución a la problemática de generar presentaciones multimedia de manera automática. Dichas presentaciones deben mostrar de una manera clara y sencilla la información a partir de planes de presentación de noticias y un sistema de base de datos. Para dicho fin se pretende desarrollar una herramienta, que genere presentaciones, a modo de periódico virtual. Las presentaciones de las noticias permiten al usuario interactuar de manera directa con los diferentes elementos de la presentación, ya sean texto, imágenes, animaciones, gráficos, mapas, etc. lo que facilita la comprensión de la información que se quiere ofrecer al usuario. De manera más detallada se presentan los objetivos concretos para el desarrollo del presente proyecto fin de carrera: • Análisis del problema. Se realiza un estudio de la problemática que se pretende solucionar con el sistema. Se revisan las técnicas y estado de la técnica en el campo Generador de interfaces de usuario como periódicos gráficos interactivos 4 de la generación automática de interfaces de usuario y su adaptación para ofrecer la difusión de la información a través de internet mediante aplicaciones web. Se identifica y estudia el contexto del sistema, el cual se orienta a la simulación del periodismo de datos. • Diseño del sistema. Se pretende diseñar un sistema informático completo, desde la capa de datos hasta la capa de presentación, que cumpla con los objetivos o requisitos propuestos. El diseño es general dado que un objetivo es permitir que el sistema pueda operar en dominios de naturaleza variada. • Implementación. Se aborda la fase de implementación con el objetivo de realizar un código modular, que sea fácil de mantener y aumentar, ya que esta herramienta pretende ofrecer servicio a diferentes sistemas de dominios variados y por lo tanto pueden surgir nuevas necesidades según el ámbito donde opere. Otro objetivo en esta parte del desarrollo, es que el sistema sea eficiente y capaz de manejar grandes volúmenes de datos así como realizar una adecuada generación automática de la presentación multimedia con texto y gráficos. • Validación. Para verificar el buen funcionamiento del sistema se desarrollan pruebas específicas que prueban las funcionalidades del sistema de manera unitaria. Para probar la generalidad del sistema se plantean diversos ejemplos de noticias en diferentes escenarios o dominios de datos, lo que nos permite comprobar la versatilidad del sistema. Para probar la robustez y solidez del sistema se pretende introducir este generador de interfaces de usuario dentro de un sistema generador automático de noticias de información hidrológica en tiempo real. El sistema generador de interfaces de usuario debe recoger cada plan de noticia, realizar automáticamente la presentación de cada noticia y organizar toda la colección de dichas noticias en un periódico virtual adaptado a las noticias que va a presentar. ORGANIZACIÓN DE LA MEMORIA La memoria se estructura de la siguiente manera; en primer lugar se presenta el estado de la técnica, con los avances y últimos desarrollos en la temática de la generación automática de interfaces de usuario, las aplicaciones web como solución de interfaz de usuario y de difusión a través de internet y los avances en simulación de periodismo dirigido por datos. Seguidamente se presenta la descripción del software, con una explicación de las diferentes fases del proceso de desarrollo software, como análisis, diseño e implementación, así como la ubicación que ocupa este sistema en un sistema de simulación de periodismo de datos. En esta parte se muestran los detalles concretos a tecnologías, lenguajes y bibliotecas utilizadas. En el siguiente capítulo se trata la validación del producto, se detallan las características del sistema automático de información hidrológica (SAIH), con el cual se ha podido probar completamente el sistema de generación automática de interfaces de usuario. Seguidamente se detallan las conclusiones a las que se ha llegado después de la realización de este proyecto, haciendo un repaso de los objetivos planteados así como las líneas futuras sobre estas técnicas. Por último se adjunta el manual de usuario de la herramienta y una colección de ejemplos mostrando las entradas y salidas procesadas por el software.

Designing a high performance parallel logic programming system

Compilation techniques such as those portrayed by the Warren Abstract Machine(WAM) have greatly improved the speed of execution of logic programs. The research presented herein is geared towards providing additional performance to logic programs through the use of parallelism, while preserving the conventional semantics of logic languages. Two áreas to which special attention is given are the preservation of sequential performance and storage efficiency, and the use of low overhead mechanisms for controlling parallel execution. Accordingly, the techniques used for supporting parallelism are efficient extensions of those which have brought high inferencing speeds to sequential implementations. At a lower level, special attention is also given to design and simulation detail and to the architectural implications of the execution model behavior. This paper offers an overview of the basic concepts and techniques used in the parallel design, simulation tools used, and some of the results obtained to date.

Effectiveness of abstract interpretation in automatic parallelization: a case study in logic programming

We report on a detailed study of the application and effectiveness of program analysis based on abstract interpretation to automatic program parallelization. We study the case of parallelizing logic programs using the notion of strict independence. We first propose and prove correct a methodology for the application in the parallelization task of the information inferred by abstract interpretation, using a parametric domain. The methodology is generic in the sense of allowing the use of different analysis domains. A number of well-known approximation domains are then studied and the transformation into the parametric domain defined. The transformation directly illustrates the relevance and applicability of each abstract domain for the application. Both local and global analyzers are then built using these domains and embedded in a complete parallelizing compiler. Then, the performance of the domains in this context is assessed through a number of experiments. A comparatively wide range of aspects is studied, from the resources needed by the analyzers in terms of time and memory to the actual benefits obtained from the information inferred. Such benefits are evaluated both in terms of the characteristics of the parallelized code and of the actual speedups obtained from it. The results show that data flow analysis plays an important role in achieving efficient parallelizations, and that the cost of such analysis can be reasonable even for quite sophisticated abstract domains. Furthermore, the results also offer significant insight into the characteristics of the domains, the demands of the application, and the trade-offs involved.

Digital signage application : developing a public venue multimedia touch-screen

La señalización digital o digital signage es una tecnología de comunicaciones digital que se está usando en los últimos años para reemplazar a la antigua publicidad impresa. Esta tecnología mejora la presentación y promoción de los productos anunciados, así como facilita el intercambio de información gracias a su colocación en lugares públicos o al aire libre. Las aplicaciones con las que cuenta este nuevo método de publicidad son muy variadas, ya que pueden variar desde ambientes privados en empresas, hasta lugares públicos como centros comerciales. Aunque la primera y principal utilidad de la señalización digital es la publicidad para que el usuario sienta una necesidad de adquirir productos, también la posibilidad de ofrecer más información sobre determinados artículos a través de las nuevas tecnologías es muy importante en este campo. La aplicación realizada en este proyecto es el desarrollo de un programa en Adobe Flash a través de lenguaje de programación XML. A través de una pantalla táctil, el usuario de un museo puede interactivamente acceder a un menú en el que aparecen los diferentes estilos de arte en un determinado tiempo de la historia. A través de una línea de tiempo se puede acceder a información sobre cada objeto que esté expuesto en la exhibición. Además se pueden observar imágenes de los detalles más importantes del objeto que no pueden ser vistos a simple vista, ya que no está permitido manipularlos. El empleo de la pantalla interactiva sirve para el usuario de la exhibición como una herramienta extra para recabar información sobre lo que está viendo, a través de una tecnología nueva y fácil de usar para todo el mundo, ya que solo se necesita usar las propias manos. La facilidad de manejo en aplicaciones como estas es muy importante, ya que el usuario final puede no tener conocimientos tecnológicos por lo que la información debe darse claramente. Como conclusión, se puede decir que digital signage es un mercado que está en expansión y que las empresas deben invertir en el desarrollo de contenidos, ya que las tecnologías avanzan aunque el digital signage no lo haga, y este sector podría ser muy útil en un futuro no muy lejano, ya que la información que es capaz de transmitir al espectador en todos los lugares es mucho más válida y útil que la proporcionada por un simple póster impreso en una valla publicitaria. Abstract The Digital signage is a digital communications technology being used in recent years to replace the old advertising printed. This technology improves the presentation and promotion of the advertised products, and makes easy the exchange of information with its placement in public places or outdoors. The applications that account this new method of advertising are several; they can range from private rooms in companies, to public places like malls. Although the first major utility of Digital signage is the advertising that makes the user feel and need of purchasing products. In addition, the chance of providing more information about certain items through new technologies is very important in this field. The application made in this project is the development of a program in Adobe Flash via XML programming language. Through a touch-screen, a museum user can interactively access a menu in which different styles of art in a particular time in history appears. Through a timeline you can access to information about each object that is exposed on display. Also you can see pictures of the most important details of the object that can not be seen with the naked eye, since it is not allowed to manipulate it. The use of the interactive screen serves to the user exhibition as an extra tool to gather information about what is seeing through a new technology and easy to use for everyone, since only need to use one’s own hands. The ease of handling in applications such as this is very important as the end user may not have expertise technology so the information should be clearly. As conclusion, one can say digital signage is an expansion market and companies must invest in content development, as although digital technologies advance digital signage does not, and this sector could be very useful in a near future, because information that is able of transmitting the everywhere viewer is much more valid and useful than that provided by a simple printed poster on a billboard.

Distributed www programming using (Ciao-) prolog and the pillow library

We discuss from a practical point of view a number of ssues involved in writing distributed Internet and WWW applications using LP/CLP systems. We describe PiLLoW, a publicdomain Internet and WWW programming library for LP/CLP systems that we have designed in order to simplify the process of writing such applications. PiLLoW provides facilities for accessing documents and code on the WWW; parsing, manipulating and generating HTML and XML structured documents and data; producing HTML forms; writing form handlers and CGI-scripts; and processing HTML/XML templates. An important contribution of PÍ'LLOW is to model HTML/XML code (and, thus, the content of WWW pages) as terms. The PÍ'LLOW library has been developed in the context of the Ciao Prolog system, but it has been adapted to a number of popular LP/CLP systems, supporting most of its functionality. We also describe the use of concurrency and a highlevel model of client-server interaction, Ciao Prolog's active modules, in the context of WWW programming. We propose a solution for client-side downloading and execution of Prolog code, using generic browsers. Finally, we also provide an overview of related work on the topic.

Abstract multiple specialization and its application to program parallelization.

Program specialization optimizes programs for known valúes of the input. It is often the case that the set of possible input valúes is unknown, or this set is infinite. However, a form of specialization can still be performed in such cases by means of abstract interpretation, specialization then being with respect to abstract valúes (substitutions), rather than concrete ones. We study the múltiple specialization of logic programs based on abstract interpretation. This involves in principie, and based on information from global analysis, generating several versions of a program predicate for different uses of such predicate, optimizing these versions, and, finally, producing a new, "multiply specialized" program. While múltiple specialization has received theoretical attention, little previous evidence exists on its practicality. In this paper we report on the incorporation of múltiple specialization in a parallelizing compiler and quantify its effects. A novel approach to the design and implementation of the specialization system is proposed. The resulting implementation techniques result in identical specializations to those of the best previously proposed techniques but require little or no modification of some existing abstract interpreters. Our results show that, using the proposed techniques, the resulting "abstract múltiple specialization" is indeed a relevant technique in practice. In particular, in the parallelizing compiler application, a good number of run-time tests are eliminated and invariants extracted automatically from loops, resulting generally in lower overheads and in several cases in increased speedups.

Some challenges for constraint programming

We propose a number of challenges for future constraint programming systems, including improvements in implementation technology (using global analysis based optimization and parallelism), debugging facilities, and the extensión of the application domain to distributed, global programming. We also briefly discuss how we are exploring techniques to meet these challenges in the context of the development of the CIAO constraint logic programming system.

Strategic directions in constraint programming

An abstract is not available.

A generator of efficient abstract machine implementations and its application to emulator minimization

The implementation of abstract machines involves complex decisions regarding, e.g., data representation, opcodes, or instruction specialization levéis, all of which affect the final performance of the emulator and the size of the bytecode programs in ways that are often difficult to foresee. Besides, studying alternatives by implementing abstract machine variants is a time-consuming and error-prone task because of the level of complexity and optimization of competitive implementations, which makes them generally difficult to understand, maintain, and modify. This also makes it hard to genérate specific implementations for particular purposes. To ameliorate those problems, we propose a systematic approach to the automatic generation of implementations of abstract machines. Different parts of their definition (e.g., the instruction set or the infernal data and bytecode representation) are kept sepárate and automatically assembled in the generation process. Alternative versions of the abstract machine are therefore easier to produce, and variants of their implementation can be created mechanically, with specific characteristics for a particular application if necessary. We illustrate the practicality of the approach by reporting on an implementation of a generator of production-quality WAMs which are specialized for executing a particular fixed (set of) program(s). The experimental results show that the approach is effective in reducing emulator size.

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.

A framework for assertion-based debugging in constraint logic programming

We propose a general framework for assertion-based debugging of constraint logic programs. Assertions are linguistic constructions which allow expressing properties of programs. We define assertion schemas which allow writing (partial) specifications for constraint logic programs using quite general properties, including user-defined programs. The framework is aimed at detecting deviations of the program behavior (symptoms) with respect to the given assertions, either at compile-time or run-time. We provide techniques for using information from global analysis both to detect at compile-time assertions which do not hold in at least one of the possible executions (i.e., static symptoms) and assertions which hold for all possible executions (i.e., statically proved assertions). We also provide program transformations which introduce tests in the program for checking at run-time those assertions whose status cannot be determined at compile-time. Both the static and the dynamic checking are provably safe in the sense that all errors flagged are definite violations of the specifications. Finally, we report on an implemented instance of the assertion language and framework.

IDRA (IDeal Resource Allocation): Computing ideal speedups in parallel logic programming

We present a technique to estimate accurate speedups for parallel logic programs with relative independence from characteristics of a given implementation or underlying parallel hardware. The proposed technique is based on gathering accurate data describing one execution at run-time, which is fed to a simulator. Alternative schedulings are then simulated and estimates computed for the corresponding speedups. A tool implementing the aforementioned techniques is presented, and its predictions are compared to the performance of real systems, showing good correlation.

Using attributed variables in the implementation of concurrent and parallel logic programming systems

Incorporating the possibility of attaching attributes to variables in a logic programming system has been shown to allow the addition of general constraint solving capabilities to it. This approach is very attractive in that by adding a few primitives any logic programming system can be turned into a generic constraint logic programming system in which constraint solving can be user deñned, and at source level - an extreme example of the "glass box" approach. In this paper we propose a different and novel use for the concept of attributed variables: developing a generic parallel/concurrent (constraint) logic programming system, using the same "glass box" flavor. We argüe that a system which implements attributed variables and a few additional primitives can be easily customized at source level to implement many of the languages and execution models of parallelism and concurrency currently proposed, in both shared memory and distributed systems. We illustrate this through examples and report on an implementation of our ideas.

Towards a concurrent semantics-based analysis of CC and CLP. principles and practice of constraint programming

Abstract is not available.