Raúl Monge

Building a security reference architecture for cloud systems

Reference architectures (RAs) are useful tools to understand and build complex systems, and many cloud providers and software product vendors have developed versions of them. RAs describe at an abstract level (no implementation details) the main features of their cloud systems. Security is a fundamental concern in clouds and several cloud vendors provide security reference architectures (SRAs) to describe the security features of their services. A SRA is an abstract architecture describing a conceptual model of security for a cloud system and provides a way to specify security requirements for a wide range of concrete architectures. We propose here a method to build a SRA for clouds defined using UML models and patterns, which goes beyond existing models in providing a global view and a more precise description. We present a metamodel as well as security and misuse patterns for this purpose. We validate our approach by showing that it can describe more precisely existing models and that it has a variety of uses. We describe in detail one of these uses, a way of evaluating the security level of a SRA.

Parallel Approach for Ensemble Learning with Locally Coupled Neural Networks

Ensemble learning has gained considerable attention in different tasks including regression, classification and clustering. Adaboost and Bagging are two popular approaches used to train these models. The former provides accurate estimations in regression settings but is computationally expensive because of its inherently sequential structure, while the latter is less accurate but highly efficient. One of the drawbacks of the ensemble algorithms is the high computational cost of the training stage. To address this issue, we propose a parallel implementation of the Resampling Local Negative Correlation (RLNC) algorithm for training a neural network ensemble in order to acquire a competitive accuracy like that of Adaboost and an efficiency comparable to that of Bagging. We test our approach on both synthetic and real datasets from the UCI and Statlib repositories for the regression task. In particular, our fine-grained parallel approach allows us to achieve a satisfactory balance between accuracy and parallel efficiency.

Context-Aware Regression from Distributed Sources

In this paper we present a distributed regression framework to model data with different contexts. Different context is defined as the change of the underlying laws of probability in the distributed sources. Most state of the art methods do not take into account the different context and assume that the data comes from the same statistical distribution. We propose an aggregation scheme for models that are in the same neighborhood in terms of statistical divergence.We conduct experiments with synthetic data sets to validate our proposal. Our proposed algorithm outperforms other models that follow a traditional approach.

Wind Speed Forecast under a Distributed Learning Approach

In this paper we apply a distributed learning approach to improve the perfomance of wind speed forecast. We use data obtained from 54 different weather stations in the U. S. and without sharing data between sites, we share model information between them, to improve the performance over local models trained with only local data. We show that sharing the information of the distributed models, improves the forecast we could obtain by only using locally trained models.

A reference architecture for web browsers: part II, a pattern for web browser content renderer

Currently, most software developments are focused in creating systems connected to the Internet, which allows to add functionality within a system and facilities to their stakeholders. This leads to depend on a web client, such as a web browser, which allows access to Internet services, data or operations that a system delivers. Within the browsers main components, a rendering engine is in charge of obtaining a convenient data structure as an output for a browser process. We developed a Web Browser Communication pattern that describes the infrastructure to allow the communication between a web client (or a web browser) and a server in the Internet. In this paper, we describe the component in charge of the rendering for a obtained web resource within the web browser, named here as Web Browser Content Renderer. In this work we have described this component as a pattern, to describe how a browsers rendering engine works and interacts with other subsystems. Patterns combine experience and good practices to obtain models that can be used for new designs, to compare and select systems/applications or to teach others. The audience to which our paper is focused are browser developers, web application developers, researchers and teachers, being the first two the most important.

Regression from Distributed Data Sources Using Discrete Neighborhood Representations and Modified Stalked Generalization Models

In this work we present a Distributed Regression approach, which works in problems where distributed data sources may have different contexts. Different context is defined as the change of the underlying law of probability in the distributed sources. We present an approach which uses a discrete representation of the probability density functions (pdfs). We create neighborhoods of similar datasets, comparing their pdfs, and use this information to build an ensemble-based approach and to improve a second level model used in this proposal, that is based in stalked generalization. We compare the proposal with other state of the art models with 5 real data sets and obtain favorable results in the majority of the datasets.

A security reference architecture for cloud systems

Security is a fundamental concern in clouds and several cloud vendors provide Security Reference Architectures (SRAs) to describe the security level of their services. A SRA is an abstract architecture without implementation details showing a conceptual model of security for a cloud system. In general, Reference Architectures (RAs) are becoming useful tools to understand and build complex systems. We propose here a Security Reference Architecture (SRA), defined using UML models and patterns, incorporating a specific approach to build secure systems. We present a metamodel and possible patterns to conceptualize the approach. We also describe some uses for this SRA, including its value for Service Level Agreements (SLAs), service certification, monitoring, and security evaluation. We show this latter use in some detail.