A uniform approach for programming distributed heterogeneous computing systems

Large-scale compute clusters of heterogeneous nodes equipped with multi-core CPUs and GPUs are getting increasingly popular in the scientific community. However, such systems require a combination of different programming paradigms making application development very challenging.In this article we introduce libWater, a library-based extension of the OpenCL programming model that simplifies the development of heterogeneous distributed applications. libWater consists of a simple interface, which is a transparent abstraction of the underlying distributed architecture, offering advanced features such as inter-context and inter-node device synchronization. It provides a runtime system which tracks dependency information enforced by event synchronization to dynamically build a DAG of commands, on which we automatically apply two optimizations: collective communication pattern detection and device-host-device copy removal.We assess libWater’s performance in three compute clusters available from the Vienna Scientific Cluster, the Barcelona Supercomputing Center and the University of Innsbruck, demonstrating improved performance and scaling with different test applications and configurations.     

Recommending suitable learning scenarios according to learners’ preferences: An improved swarm based approach

The paper presents a new approach for recommending suitable learning paths for different learners groups. Selection of the learning path is considered as recommendations to choosing and combining the sequences of learning objects (LOs) according to learners’ preferences. Learning path can be selected by applying artificial intelligence techniques, e.g. a swarm intelligence model. If we modify and/or change some LOs in the learning path, we should rearrange the alignment of new and old LOs and reallocate pheromones to achieve effective learning recommendations. To solve this problem, a new method based on the ant colony optimisation algorithm and adaptation of the solution to the changing optimum is proposed. A simulation process with a dynamic change of learning paths when new LOs are inserted was chosen to verify the method proposed. The paper contributes with the following new developments: (1) an approach of dynamic learning paths selection based on swarm intelligence, and (2) a modified ant colony optimisation algorithm for learning paths selection. The elaborated approach effectively assist learners by helping them to reach most suitable LOs according to their preferences, and tutors – by helping them to monitor, refine, and improve e-learning modules and courses according to the learners’ behaviour.     

A Dantzig–Wolfe decomposition algorithm for linear economic model predictive control of dynamically decoupled subsystems

This paper presents a warm-started Dantzig–Wolfe decomposition algorithm tailored to economic model predictive control of dynamically decoupled subsystems. We formulate the constrained optimal control problem solved at each sampling instant as a linear program with state space constraints, input limits, input rate limits, and soft output limits. The objective function of the linear program is related directly to the cost of operating the subsystems, and the cost of violating the soft output constraints. Simulations for large-scale economic power dispatch problems show that the proposed algorithm is significantly faster than both state-of-the-art linear programming solvers, and a structure exploiting implementation of the alternating direction method of multipliers. It is also demonstrated that the control strategy presented in this paper can be tuned using a weighted ℓ₁-regularization term. In the presence of process and measurement noise, such a regularization term is critical for achieving a well-behaved closed-loop performance.     

Profit optimisation of the multiple-vacation machine repair problem using particle swarm optimisation

This paper investigates a multiple-vacation M/M/1 warm-standby machine repair problem with an unreliable repairman. We first apply a matrix-analytic method to obtain the steady-state probabilities. Next, we construct the total expected profit per unit time and formulate an optimisation problem to find the maximum profit. The particle swarm optimisation (PSO) algorithm is implemented to determine the optimal number of warm standbys S* and the service rate μ* as well as vacation rate ν* simultaneously at the optimal maximum profit. We compare the searching results of the PSO algorithm with those of exhaustive search method to ensure the searching quality of the PSO algorithm. Sensitivity analysis with numerical illustrations is also provided.     

Ellipsoidal neighbourhood outlier factor for distributed anomaly detection in resource constrained networks

Anomaly detection in resource constrained wireless networks is an important challenge for tasks such as intrusion detection, quality assurance and event monitoring applications. The challenge is to detect these interesting events or anomalies in a timely manner, while minimising energy consumption in the network. We propose a distributed anomaly detection architecture, which uses multiple hyperellipsoidal clusters to model the data at each sensor node, and identify global and local anomalies in the network. In particular, a novel anomaly scoring method is proposed to provide a score for each hyperellipsoidal model, based on how remote the ellipsoid is relative to their neighbours. We demonstrate using several synthetic and real datasets that our proposed scheme achieves a higher detection performance with a significant reduction in communication overhead in the network compared to centralised and existing schemes.     

Fast fingerprint identification for large databases

Fingerprint matching has emerged as an effective tool for human recognition due to the uniqueness, universality and invariability of fingerprints. Many different approaches have been proposed in the literature to determine faithfully if two fingerprint images belong to the same person. Among them, minutiae-based matchers highlight as the most relevant techniques because of their discriminative capabilities, providing precise results. However, performing a fingerprint identification over a large database can be an inefficient task due to the lack of scalability and high computing times of fingerprint matching algorithms.In this paper, we propose a distributed framework for fingerprint matching to tackle large databases in a reasonable time. It provides a general scheme for any kind of matcher, so that its precision is preserved and its time of response can be reduced.To test the proposed system, we conduct an extensive study that involves both synthetic and captured fingerprint databases, which have different characteristics, analyzing the performance of three well-known minutiae-based matchers within the designed framework. With the available hardware resources, our distributed model is able to address up to 400 000 fingerprints in approximately half a second. Additional details are provided at http://sci2s.ugr.es/ParallelMatching.     

A distributed solution to the network reconstruction problem

It has been recently shown in Ren et al. (2010) that by collecting noise-contaminated time series generated by a coupled-oscillator system at each node of a network, it is possible to robustly reconstruct its topology, i.e. determine the graph Laplacian. Restricting ourselves to linear consensus dynamics over undirected communication networks, in this paper we introduce a new dynamic average consensus least-squares algorithm to locally estimate these time series at each node, thus making the reconstruction process fully distributed and more easily applicable in the real world. We also propose a novel efficient method for separating the off-diagonal entries of the reconstructed Laplacian, and examine several concepts related to the trace of the dynamic correlation matrix of the coupled single integrators, which is a distinctive element of our network reconstruction method. The theory is illustrated with examples from computer, power and transportation systems.     

Analysing interactive devices based on information resource constraints

Analysis of the usability of an interactive system requires both an understanding of how the system is to be used and a means of assessing the system against that understanding. Such analytic assessments are particularly important in safety-critical systems as latent vulnerabilities may exist which have negative consequences only in certain circumstances. Many existing approaches to assessment use tasks or scenarios to provide explicit representation of their understanding of use. These normative user behaviours have the advantage that they clarify assumptions about how the system will be used but have the disadvantage that they may exclude many plausible deviations from these norms. Assessments of how a design fails to support these user behaviours can be a matter of judgement based on individual experience rather than evidence. We present a systematic formal method for analysing interactive systems that is based on constraints rather than prescribed behaviour. These constraints capture precise assumptions about what information resources are used to perform action. These resources may either reside in the system itself or be external to the system. The approach is applied to two different medical device designs, comparing two infusion pumps currently in common use in hospitals. Comparison of the two devices is based on these resource assumptions to assess consistency of interaction within the design of each device.     

A comprehensive approach to evaluating watershed models for predicting river flow regimes critical to downstream ecosystem services

Selection of strategies that help reduce riverine inputs requires numerical models that accurately quantify hydrologic processes. While numerous models exist, information on how to evaluate and select the most robust models is limited. Toward this end, we developed a comprehensive approach that helps evaluate watershed models in their ability to simulate flow regimes critical to downstream ecosystem services. We demonstrated the method using the Soil and Water Assessment Tool (SWAT), the Hydrological Simulation Program–FORTRAN (HSPF) model, and Distributed Large Basin Runoff Model (DLBRM) applied to the Maumee River Basin (USA). The approach helped in identifying that each model simulated flows within acceptable ranges. However, each was limited in its ability to simulate flows triggered by extreme weather events, owing to algorithms not being optimized for such events and mismatched physiographic watershed conditions. Ultimately, we found HSPF to best predict river flow, whereas SWAT offered the most flexibility for evaluating agricultural management practices.