Towards granularity based control of parallelism in logic programs

Several types of parallelism can be exploited in logic programs while preserving correctness and efficiency, i.e. ensuring that the parallel execution obtains the same results as the sequential one and the amount of work performed is not greater. However, such results do not take into account a number of overheads which appear in practice, such as process creation and scheduling, which can induce a slow-down, or, at least, limit speedup, if they are not controlled in some way. This paper describes a methodology whereby the granularity of parallel tasks, i.e. the work available under them, is efficiently estimated and used to limit parallelism so that the effect of such overheads is controlled. The run-time overhead associated with the approach is usually quite small, since as much work is done at compile time as possible. Also, a number of run-time optimizations are proposed. Moreover, a static analysis of the overhead associated with the granularity control process is performed in order to decide its convenience. The performance improvements resulting from the incorporation of grain size control are shown to be quite good, specially for systems with médium to large parallel execution overheads.

Interval-based resource usage verification: Formalization and prototype

In an increasing number of applications (e.g., in embedded, real-time, or mobile systems) it is important or even essential to ensure conformance with respect to a specification expressing resource usages, such as execution time, memory, energy, or user-defined resources. In previous work we have presented a novel framework for data size-aware, static resource usage verification. Specifications can include both lower and upper bound resource usage functions. In order to statically check such specifications, both upper- and lower-bound resource usage functions (on input data sizes) approximating the actual resource usage of the program which are automatically inferred and compared against the specification. The outcome of the static checking of assertions can express intervals for the input data sizes such that a given specification can be proved for some intervals but disproved for others. After an overview of the approach in this paper we provide a number of novel contributions: we present a full formalization, and we report on and provide results from an implementation within the Ciao/CiaoPP framework (which provides a general, unified platform for static and run-time verification, as well as unit testing). We also generalize the checking of assertions to allow preconditions expressing intervals within which the input data size of a program is supposed to lie (i.e., intervals for which each assertion is applicable), and we extend the class of resource usage functions that can be checked.

Functional notation and lazy evaluation in Ciao

Certain aspects of functional programming provide syntactic convenience, such as having a designated implicit output argument, which allows function cali nesting and sometimes results in more compact code. Functional programming also sometimes allows a more direct encoding of lazy evaluation, with its ability to deal with infinite data structures. We present a syntactic functional extensión of Prolog covering function application, predefined evaluable functors, functional definitions, quoting, and lazy evaluation. The extensión is also composable with higher-order features. We also highlight the Ciao features which help implementation and present some data on the overhead of using lazy evaluation with respect to eager evaluation.

Análisis estático de sistemas de membranas para la determinación de reglas activas

Esta tesis doctoral se enmarca dentro de la computación con membranas. Se trata de un tipo de computación bio-inspirado, concretamente basado en las células de los organismos vivos, en las que se producen múltiples reacciones de forma simultánea. A partir de la estructura y funcionamiento de las células se han definido diferentes modelos formales, denominados P sistemas. Estos modelos no tratan de modelar el comportamiento biológico de una célula, sino que abstraen sus principios básicos con objeto de encontrar nuevos paradigmas computacionales. Los P sistemas son modelos de computación no deterministas y masivamente paralelos. De ahí el interés que en los últimos años estos modelos han suscitado para la resolución de problemas complejos. En muchos casos, consiguen resolver de forma teórica problemas NP-completos en tiempo polinómico o lineal. Por otra parte, cabe destacar también la aplicación que la computación con membranas ha tenido en la investigación de otros muchos campos, sobre todo relacionados con la biología. Actualmente, una gran cantidad de estos modelos de computación han sido estudiados desde el punto de vista teórico. Sin embargo, el modo en que pueden ser implementados es un reto de investigación todavía abierto. Existen varias líneas en este sentido, basadas en arquitecturas distribuidas o en hardware dedicado, que pretenden acercarse en lo posible a su carácter no determinista y masivamente paralelo, dentro de un contexto de viabilidad y eficiencia. En esta tesis doctoral se propone la realización de un análisis estático del P sistema, como vía para optimizar la ejecución del mismo en estas plataformas. Se pretende que la información recogida en tiempo de análisis sirva para configurar adecuadamente la plataforma donde se vaya a ejecutar posteriormente el P sistema, obteniendo como consecuencia una mejora en el rendimiento. Concretamente, en esta tesis se han tomado como referencia los P sistemas de transiciones para llevar a cabo el estudio de dicho análisis estático. De manera un poco más específica, el análisis estático propuesto en esta tesis persigue que cada membrana sea capaz de determinar sus reglas activas de forma eficiente en cada paso de evolución, es decir, aquellas reglas que reúnen las condiciones adecuadas para poder ser aplicadas. En esta línea, se afronta el problema de los estados de utilidad de una membrana dada, que en tiempo de ejecución permitirán a la misma conocer en todo momento las membranas con las que puede comunicarse, cuestión que determina las reglas que pueden aplicarse en cada momento. Además, el análisis estático propuesto en esta tesis se basa en otra serie de características del P sistema como la estructura de membranas, antecedentes de las reglas, consecuentes de las reglas o prioridades. Una vez obtenida toda esta información en tiempo de análisis, se estructura en forma de árbol de decisión, con objeto de que en tiempo de ejecución la membrana obtenga las reglas activas de la forma más eficiente posible. Por otra parte, en esta tesis se lleva a cabo un recorrido por un número importante de arquitecturas hardware y software que diferentes autores han propuesto para implementar P sistemas. Fundamentalmente, arquitecturas distribuidas, hardware dedicado basado en tarjetas FPGA y plataformas basadas en microcontroladores PIC. El objetivo es proponer soluciones que permitan implantar en dichas arquitecturas los resultados obtenidos del análisis estático (estados de utilidad y árboles de decisión para reglas activas). En líneas generales, se obtienen conclusiones positivas, en el sentido de que dichas optimizaciones se integran adecuadamente en las arquitecturas sin penalizaciones significativas. Summary Membrane computing is the focus of this doctoral thesis. It can be considered a bio-inspired computing type. Specifically, it is based on living cells, in which many reactions take place simultaneously. From cell structure and operation, many different formal models have been defined, named P systems. These models do not try to model the biological behavior of the cell, but they abstract the basic principles of the cell in order to find out new computational paradigms. P systems are non-deterministic and massively parallel computational models. This is why, they have aroused interest when dealing with complex problems nowadays. In many cases, they manage to solve in theory NP problems in polynomial or lineal time. On the other hand, it is important to note that membrane computing has been successfully applied in many researching areas, specially related to biology. Nowadays, lots of these computing models have been sufficiently characterized from a theoretical point of view. However, the way in which they can be implemented is a research challenge, that it is still open nowadays. There are some lines in this way, based on distributed architectures or dedicated hardware. All of them are trying to approach to its non-deterministic and parallel character as much as possible, taking into account viability and efficiency. In this doctoral thesis it is proposed carrying out a static analysis of the P system in order to optimize its performance in a computing platform. The general idea is that after data are collected in analysis time, they are used for getting a suitable configuration of the computing platform in which P system is going to be performed. As a consequence, the system throughput will improve. Specifically, this thesis has made use of Transition P systems for carrying out the study in static analysis. In particular, the static analysis proposed in this doctoral thesis tries to achieve that every membrane can efficiently determine its active rules in every evolution step. These rules are the ones that can be applied depending on the system configuration at each computational step. In this line, we are going to tackle the problem of the usefulness states for a membrane. This state will allow this membrane to know the set of membranes with which communication is possible at any time. This is a very important issue in determining the set of rules that can be applied. Moreover, static analysis in this thesis is carried out taking into account other properties such as membrane structure, rule antecedents, rule consequents and priorities among rules. After collecting all data in analysis time, they are arranged in a decision tree structure, enabling membranes to obtain the set of active rules as efficiently as possible in run-time system. On the other hand, in this doctoral thesis is going to carry out an overview of hardware and software architectures, proposed by different authors in order to implement P systems, such as distributed architectures, dedicated hardware based on PFGA, and computing platforms based on PIC microcontrollers. The aim of this overview is to propose solutions for implementing the results of the static analysis, that is, usefulness states and decision trees for active rules. In general, conclusions are satisfactory, because these optimizations can be properly integrated in most of the architectures without significant penalties.

A simulation study on parallel backtracking with solution memoing for independent and-parallelism

Goal-level Independent and-parallelism (IAP) is exploited by scheduling for simultaneous execution two or more goals which will not interfere with each other at run time. This can be done safely even if such goals can produce multiple answers. The most successful IAP implementations to date have used recomputation of answers and sequentially ordered backtracking. While in principle simplifying the implementation, recomputation can be very inefficient if the granularity of the parallel goals is large enough and they produce several answers, while sequentially ordered backtracking limits parallelism. And, despite the expected simplification, the implementation of the classic schemes has proved to involve complex engineering, with the consequent difficulty for system maintenance and expansion, and still frequently run into the well-known trapped goal and garbage slot problems. This work presents ideas about an alternative parallel backtracking model for IAP and a simulation studio. The model features parallel out-of-order backtracking and relies on answer memoization to reuse and combine answers. Whenever a parallel goal backtracks, its siblings also perform backtracking, but after storing the bindings generated by previous answers. The bindings are then reinstalled when combining answers. In order not to unnecessarily penalize forward execution, non-speculative and-parallel goals which have not been executed yet take precedence over sibling goals which could be backtracked over. Using a simulator, we show that this approach can bring significant performance advantages over classical approaches.

Precise set sharin analysis for java-style programs (and proofs).

Finding useful sharing information between instances in object- oriented programs has recently been the focus of much research. The applications of such static analysis are multiple: by knowing which variables definitely do not share in memory we can apply conventional compiler optimizations, find coarse-grained parallelism opportunities, or, more importantly, verify certain correctness aspects of programs even in the absence of annotations. In this paper we introduce a framework for deriving precise sharing information based on abstract interpretation for a Java-like language. Our analysis achieves precision in various ways, including supporting multivariance, which allows separating different contexts. We propose a combined Set Sharing + Nullity + Classes domain which captures which instances do not share and which ones are definitively null, and which uses the classes to refine the static information when inheritance is present. The use of a set sharing abstraction allows a more precise representation of the existing sharings and is crucial in achieving precision during interprocedural analysis. Carrying the domains in a combined way facilitates the interaction among them in the presence of multivariance in the analysis. We show through examples and experimentally that both the set sharing part of the domain as well as the combined domain provide more accurate information than previous work based on pair sharing domains, at reasonable cost.

Precise set sharing and nullity analysis for java-style program

Finding useful sharing information between instances in object- oriented programs has been recently the focus of much research. The applications of such static analysis are multiple: by knowing which variables share in memory we can apply conventional compiler optimizations, find coarse-grained parallelism opportunities, or, more importantly,erify certain correctness aspects of programs even in the absence of annotations In this paper we introduce a framework for deriving precise sharing information based on abstract interpretation for a Java-like language. Our analysis achieves precision in various ways. The analysis is multivariant, which allows separating different contexts. We propose a combined Set Sharing + Nullity + Classes domain which captures which instances share and which ones do not or are definitively null, and which uses the classes to refine the static information when inheritance is present. Carrying the domains in a combined way facilitates the interaction among the domains in the presence of mutivariance in the analysis. We show that both the set sharing part of the domain as well as the combined domain provide more accurate information than previous work based on pair sharing domains, at reasonable cost.

Using global analysis, partial specifications, and an extensible assertion language for program validation and debugging

We present a framework for the application of abstract interpretation as an aid during program development, rather than in the more traditional application of program optimization. Program validation and detection of errors is first performed statically by comparing (partial) specifications written in terms of assertions against information obtained from static analysis of the program. The results of this process are expressed in the user assertion language. Assertions (or parts of assertions) which cannot be verified statically are translated into run-time tests. The framework allows the use of assertions to be optional. It also allows using very general properties in assertions, beyond the predefined set understandable by the static analyzer and including properties defined by means of user programs. We also report briefly on an implementation of the framework. The resulting tool generates and checks assertions for Prolog, CLP(R), and CHIP/CLP(fd) programs, and integrates compile-time and run-time checking in a uniform way. The tool allows using properties such as types, modes, non-failure, determinacy, and computational cost, and can treat modules separately, performing incremental analysis. In practice, this modularity allows detecting statically bugs in user programs even if they do not contain any assertions.

Specialization and optimization of constraint programs with dynamic scheduling

In this report we discuss some of the issues involved in the specialization and optimization of constraint logic programs with dynamic scheduling. Dynamic scheduling, as any other form of concurrency, increases the expressive power of constraint logic programs, but also introduces run-time overhead. The objective of the specialization and optimization is to reduce as much as possible such overhead automatically, while preserving the semantics of the original programs. This is done by program transformation based on global analysis. We present implementation techniques for this purpose and report on experimental results obtained from an implementation of the techniques in the context of the CIAO compiler.

A technique for dynamic term size computation via program transformation

Knowing the size of the terms to which program variables are bound at run-time in logic programs is required in a class of applications related to program optimization such as, for example, granularity analysis and selection among different algorithms or control rules whose performance may be dependent on such size. Such size is difficult to even approximate at compile time and is thus generally computed at run-time by using (possibly predefined) predicates which traverse the terms involved. We propose a technique based on program transformation which has the potential of performing this computation much more efficiently. The technique is based on finding program procedures which are called before those in which knowledge regarding term sizes is needed and which traverse the terms whose size is to be determined, and transforming such procedures so that they compute term sizes "on the fly". We present a systematic way of determining whether a given program can be transformed in order to compute a given term size at a given program point without additional term traversal. Also, if several such transformations are possible our approach allows finding minimal transformations under certain criteria. We also discuss the advantages and applications of our technique and present some performance results.

Dynamic term size computation in logic programs via program transformation

Knowing the size of the terms to which program variables are bound at run-time in logic programs is required in a class of applications related to program optimization such as, for example, recursion elimination and granularity analysis. Such size is difficult to even approximate at compile time and is thus generally computed at run-time by using (possibly predefined) predicates which traverse the terms involved. We propose a technique based on program transformation which has the potential of performing this computation much more efficiently. The technique is based on finding program procedures which are called before those in which knowledge regarding term sizes is needed and which traverse the terms whose size is to be determined, and transforming such procedures so that they compute term sizes "on the fly". We present a systematic way of determining whether a given program can be transformed in order to compute a given term size at a given program point without additional term traversal. Also, if several such transformations are possible our approach allows finding minimal transformations under certain criteria. We also discuss the advantages and present some applications of our technique.

Ecient Inference-aware RDB2RDF Query Rewriting

RDB2RDF systems generate RDF from relational databases, operating in two dierent manners: materializing the database content into RDF or acting as virtual RDF datastores that transform SPARQL queries into SQL. In the former, inferences on the RDF data (taking into account the ontologies that they are related to) are normally done by the RDF triple store where the RDF data is materialised and hence the results of the query answering process depend on the store. In the latter, existing RDB2RDF systems do not normally perform such inferences at query time. This paper shows how the algorithm used in the REQUIEM system, focused on handling run-time inferences for query answering, can be adapted to handle such inferences for query answering in combination with RDB2RDF systems.

Ecient Inference-aware RDB2RDF Query Rewriting

RDB2RDF systems generate RDF from relational databases, operating in two di�erent manners: materializing the database content into RDF or acting as virtual RDF datastores that transform SPARQL queries into SQL. In the former, inferences on the RDF data (taking into account the ontologies that they are related to) are normally done by the RDF triple store where the RDF data is materialised and hence the results of the query answering process depend on the store. In the latter, existing RDB2RDF systems do not normally perform such inferences at query time. This paper shows how the algorithm used in the REQUIEM system, focused on handling run-time inferences for query answering, can be adapted to handle such inferences for query answering in combination with RDB2RDF systems.

Automatic motion compensation of free breathing acquired myocardial perfusion data by using independent component analysis

Images acquired during free breathing using first-pass gadolinium-enhanced myocardial perfusion magnetic resonance imaging (MRI) exhibit a quasiperiodic motion pattern that needs to be compensated for if a further automatic analysis of the perfusion is to be executed. In this work, we present a method to compensate this movement by combining independent component analysis (ICA) and image registration: First, we use ICA and a time?frequency analysis to identify the motion and separate it from the intensity change induced by the contrast agent. Then, synthetic reference images are created by recombining all the independent components but the one related to the motion. Therefore, the resulting image series does not exhibit motion and its images have intensities similar to those of their original counterparts. Motion compensation is then achieved by using a multi-pass image registration procedure. We tested our method on 39 image series acquired from 13 patients, covering the basal, mid and apical areas of the left heart ventricle and consisting of 58 perfusion images each. We validated our method by comparing manually tracked intensity profiles of the myocardial sections to automatically generated ones before and after registration of 13 patient data sets (39 distinct slices). We compared linear, non-linear, and combined ICA based registration approaches and previously published motion compensation schemes. Considering run-time and accuracy, a two-step ICA based motion compensation scheme that first optimizes a translation and then for non-linear transformation performed best and achieves registration of the whole series in 32 ± 12 s on a recent workstation. The proposed scheme improves the Pearsons correlation coefficient between manually and automatically obtained time?intensity curves from .84 ± .19 before registration to .96 ± .06 after registration

Harmony: Towards automated self-adaptive consistency in cloud storage

In just a few years cloud computing has become a very popular paradigm and a business success story, with storage being one of the key features. To achieve high data availability, cloud storage services rely on replication. In this context, one major challenge is data consistency. In contrast to traditional approaches that are mostly based on strong consistency, many cloud storage services opt for weaker consistency models in order to achieve better availability and performance. This comes at the cost of a high probability of stale data being read, as the replicas involved in the reads may not always have the most recent write. In this paper, we propose a novel approach, named Harmony, which adaptively tunes the consistency level at run-time according to the application requirements. The key idea behind Harmony is an intelligent estimation model of stale reads, allowing to elastically scale up or down the number of replicas involved in read operations to maintain a low (possibly zero) tolerable fraction of stale reads. As a result, Harmony can meet the desired consistency of the applications while achieving good performance. We have implemented Harmony and performed extensive evaluations with the Cassandra cloud storage on Grid?5000 testbed and on Amazon EC2. The results show that Harmony can achieve good performance without exceeding the tolerated number of stale reads. For instance, in contrast to the static eventual consistency used in Cassandra, Harmony reduces the stale data being read by almost 80% while adding only minimal latency. Meanwhile, it improves the throughput of the system by 45% while maintaining the desired consistency requirements of the applications when compared to the strong consistency model in Cassandra.