A scalable organization for distributed directories

Although directory-based cache-coherence protocols are the best choice when designing chip multiprocessors with tens of cores on-chip, the memory overhead introduced by the directory structure may not scale gracefully with the number of cores. Many approaches aimed at improving the scalability of directories have been proposed. However, they do not bring perfect scalability and usually reduce the directory memory overhead by compressing coherence information, which in turn results in extra unnecessary coherence messages and, therefore, wasted energy and some performance degradation. In this work, we present a distributed directory organization based on duplicate tags for tiled CMP architectures whose size is independent on the number of tiles of the system up to a certain number of tiles. We demonstrate that this number of tiles corresponds to the number of sets in the private caches. Additionally, we show that the area overhead of the proposed directory structure is 0.56% with respect to the on-chip data caches. Moreover, the proposed directory structure keeps the same information than a non-scalable full-map directory. Finally, we propose a mechanism that takes advantage of this directory organization to remove the network traffic caused by replacements. This mechanism reduces total traffic by 15% for a 16-core configuration compared to a traditional directory-based protocol.     

A multi-granular linguistic model for management decision-making in performance appraisal

The performance appraisal is a relevant process to keep and improve the competitiveness of companies in nowadays. In spite of this relevance, the current performance appraisal models are not sufficiently well-defined either designed for the evaluation framework in which they are defined. This paper proposes a performance appraisal model where the assessments are modelled by means of linguistic information provided by different sets of reviewers in order to manage the uncertainty and subjectivity of such assessments. Therefore, the reviewers could express their assessments in different linguistic scales according to their knowledge about the evaluated employees, defining a multi-granular linguistic evaluation framework. Additionally, the proposed model will manage the multi-granular linguistic labels provided by appraisers in order to compute collective assessments about the employees that will be used by the management team to make the final decision about them.     

APROMORE: An advanced process model repository

Business process models are becoming available in large numbers due to their widespread use in many industrial applications such as enterprise and quality engineering projects. On the one hand, this raises a challenge as to their proper management: how can it be ensured that the proper process model is always available to the interested stakeholder? On the other hand, the richness of a large set of process models also offers opportunities, for example with respect to the re-use of existing model parts for new models. This paper describes the functionality and architecture of an advanced process model repository, named APROMORE. This tool brings together a rich set of features for the analysis, management and usage of large sets of process models, drawing from state-of-the art research in the field of process modeling. A prototype of the platform is presented in this paper, demonstrating its feasibility, as well as an outlook on the further development of APROMORE.     

Graphical versus textual software measurement modelling: an empirical study

Model-driven Engineering (MDE) has attained great importance in both the Software Engineering industry and the research community, where it is now widely used to provide a suitable approach with which to improve productivity when developing software artefacts. In this scenario, measurement models (software artefacts) have become a fundamental point in improvement of productivity, where MDE and Software Measurement can reap mutual benefits. MDE principles and techniques can be used in software measurement to build more automatic and generic solutions, and to achieve this, it is fundamental to be able to develop software measurement models. To facilitate this task, a domain-specific language named “Software Measurement Modelling Language” (SMML) has been developed. This paper tackles the question of whether the use of SMML can assist in the definition of software measurement models. An empirical study was conducted, with the aim of verifying whether SMML makes it easier to construct measurement models which are more usable and maintainable as regards textual notation. The results show that models which do not use the language are more difficult—in terms of effort, correctness and efficiency—to understand and modify than those represented with SMML. Additional feedback was also obtained, to verify the suitability of the graphical representation of each symbol (element or relationship) of SMML.     

EDUCA: A web 2.0 authoring tool for developing adaptive and intelligent tutoring systems using a Kohonen network

This paper presents a Web 2.0 Learning Environment, for a systematic creation of adaptive and intelligent tutoring systems. Authoring contents is made by a community of users including teachers and students. The tutoring systems adapt the contents according to the best learning style using self-organizing maps (SOMs). The SOM was trained for classifying Felder–Silverman learning styles. The most important advantage of these unsupervised neural networks is that they do not require an external teacher for presenting a training set. The approach was implemented under an authoring tool that allows the production of personalized learning material to be used under collaborative and mobile learning environments. The tutoring systems together with the neural network can also be exported to mobile devices. We present different results to the approach working under the authoring tool.     

A fuzzy model to evaluate the suitability of installing an enterprise resource planning system

The use of enterprise resource planning (ERP) as a foundation for the integration of the complete range of business processes and functions, is clearly useful and economically profitable in most very large organizations which manage a great deal of data in their information systems. However, the decision of installing an ERP system in all the companies is not always so clear, it will depend on the size, future profits and other features of the company. Therefore, different parameters (features, aspects) will be evaluated to make a decision about the suitability of the ERP system. These parameters might have different nature or the knowledge about them could be vague or imprecise. Thus, this implies that it would be suitable that the evaluation process can manage heterogeneous information. In this paper we shall present a fuzzy evaluation model to evaluate the suitability of an ERP system based on a multi-expert decision-making (ME-DM) process that is able to deal with heterogeneous information.     

Air management in a diesel engine using fuzzy control techniques

Air management for diesel engines is a major challenge from the control point of view because of the highly nonlinear behavior of this system. For this reason, linear control techniques are unable to provide the required performance, and nonlinear controllers are used instead. This article discusses two fundamental steps when designing a control system. Firstly, a methodology to identify Takagi–Sugeno (T–S) structures using experimental data is proposed. Secondly, the design of a fuzzy controller in PDC structure (Parallel Distributed Compensation) is presented. The parameters of this controller are obtained from a LMI (Linear Matrix Inequalities) minimization problem.     

MOE prediction in Abies pinsapo Boiss. timber: Application of an artificial neural network using non-destructive testing

Determining the modulus of elasticity of wood by applying an artificial neural network using the physical properties and non-destructive testing can be a useful method in assessments of the timber structure in old constructions. The modulus of elasticity of Abies pinsapo Boiss. timber was predicted in this study through the parameters of density, width, thickness, moisture content, ultrasonic wave propagation velocity and visual grading of the test pieces. A feedforward multilayer perceptron network was designed for this purpose, achieving 75.0% success in the testing or unknown group.     

A study on the use of statistical tests for experimentation with neural networks: Analysis of parametric test conditions and non-parametric tests

In this paper, we focus on the experimental analysis on the performance in artificial neural networks with the use of statistical tests on the classification task. Particularly, we have studied whether the sample of results from multiple trials obtained by conventional artificial neural networks and support vector machines checks the necessary conditions for being analyzed through parametrical tests. The study is conducted by considering three possibilities on classification experiments: random variation in the selection of test data, the selection of training data and internal randomness in the learning algorithm.The results obtained state that the fulfillment of these conditions are problem-dependent and indefinite, which justifies the need of using non-parametric statistics in the experimental analysis.