Filipe Araujo

GeoPeer: a location-aware peer-to-peer system

This work presents a novel peer-to-peer system that is particularly well suited to support context-aware computing. The system, called GeoPeer, aims to combine the advantages of peer-to-peer systems that implement distributed hash tables with the suitability of geographical routing for supporting location-constrained queries and information dissemination. GeoPeer is comprised of two fundamental components: a Delaunay triangulation used to build a connected lattice of nodes and a mechanism to manage long range contacts that allows good routing performance, despite unbalanced distribution of nodes.

CHR: a distributed hash table for wireless ad hoc networks

his paper focuses on the problem of implementing a distributed hash table (DHT) in wireless ad hoc networks. Scarceness of resources and node mobility turn routing into a challenging problem and therefore, we claim that building DHT as an overlay network (like in wired environments) is not the best option. Hence, we present a proof-of-concept DHT, called cell hash routing (CHR), designed from scratch to cope with problems like limited available energy, communication range or node mobility. CHR overcomes these problems, by using position information to organize a DHT of clusters instead of individual nodes. By using position-based routing on top of these clusters, CHR is very efficient. Furthermore, its localized routing and its load sharing schemes, make CHR very scalable in respect to network size and density. For these reasons, we believe that CHR is a simple and yet powerful adaptation of the DHT concept for wireless ad hoc environments.

A neural network for shortest path computation

This paper presents a new neural network to solve the shortest path problem for inter-network routing. The proposed solution extends the traditional single-layer recurrent Hopfield architecture introducing a two-layer architecture that automatically guarantees an entire set of constraints held by any valid solution to the shortest path problem. This new method addresses some of the limitations of previous solutions, in particular the lack of reliability in what concerns successful and valid convergence. Experimental results show that an improvement in successful convergence can be achieved in certain classes of graphs. Additionally, computation performance is also improved at the expense of slightly worse results.

Characterizing result errors in internet desktop grids

Desktop grids use the free resources in Intranet and Internet environments for large-scale computation and storage. While desktop grids offer a high return on investment, one critical issue is the validation of results returned by participating hosts. Several mechanisms for result validation have been previously proposed. However, the characterization of errors is poorly understood. To study error rates, we implemented and deployed a desktop grid application across several thousand hosts distributed over the Internet. We then analyzed the results to give quantitative and empirical characterization of errors stemming from input or output (I/O) failures. We find that in practice, error rates are widespread across hosts but occur relatively infrequently. Moreover, we find that error rates tend to not be stationary over time nor correlated between hosts. In light of these characterization results, we evaluated state-of-the-art error detection mechanisms and describe the trade-offs for using each mechanism.

Scalable QoS-Based Event Routing in Publish-Subscribe Systems

This paper proposes a distributed and scalable publish-subscribe broker with support for QoS. The broker, called IndiQoS, leverages on existing mechanisms to reserve resources in the underlying network and on an overlay network of peer-to-peer rendezvous nodes, to automatically select QoS-capable paths. By avoiding flooding of either QoS reservations or link-state information, IndiQoS is able to scale with respect to network size and number of reservations. Experimental results show the validity of our approach

Classification of Systemic Lupus Erythematosus: Systemic Lupus International Collaborating Clinics Versus American College of Rheumatology Criteria. A Comparative Study of 2,055 Patients From a Real-Life, International Systemic Lupus Erythematosus Cohort.

The new Systemic Lupus International Collaborating Clinics (SLICC) 2012 classification criteria aimed to improve the performance of systemic lupus erythematosus (SLE) classification over the American College of Rheumatology (ACR) 1997 criteria. However, the SLICC 2012 criteria need further external validation. Our objective was to compare the sensitivity for SLE classification between the ACR 1997 and the SLICC 2012 criteria sets in a real‐life, multicenter, international SLE population.

On QoS-aware publish-subscribe

This paper addresses the issue of supporting quality of service (QoS) parameters in distributed publish-subscribe systems. It advocates that QoS parameters should be handled using the same constructs as other information regarding events, such as their type or content. At the same time, we claim that the use of a consistent set of mechanisms should not preclude to decouple the specification of QoS properties from the specification of type, subject or content-based constraints. We also advocate that QoS parameters should not be embedded on the type or content of the events. We show that some QoS parameters can only be computed in run-time, as they depend on dynamic aspects such as the location of the participants and the system load. The paper proposes a model that supports the decoupling of QoS characterization from the event characterization while, at the same time, offers a uniform treatment of both aspects.

Fast localized delaunay triangulation

A localized Delaunay triangulation owns the following interesting properties in a wireless ad hoc setting: it can be built with localized information, the communication cost imposed by control information is limited and it supports geographical routing algorithms that offer guaranteed convergence. This paper presents a localized algorithm that builds a graph called planar localized Delaunay triangulation, PLDel, known to be a good spanner of the unit disk graph, UDG. Unlike previous work, our algorithm builds PLDel in a single communication step, maintaining a communication cost of O(n log n), which is within a constant of the optimum. This represents a significant practical improvement over previous algorithms with similar theoretical bounds. Furthermore, the small cost of our algorithm makes feasible to use PLDel in real systems, instead of the Gabriel or the Relative Neighborhood graphs, which are not good spanners of UDG.

Prevalence of rheumatic and musculoskeletal diseases and their impact on health-related quality of life, physical function and mental health in Portugal: results from EpiReumaPt– a national health survey

Objectives To estimate the national prevalence of rheumatic and musculoskeletal diseases (RMDs) in the adult Portuguese population and to determine their impact on health-related quality of life (HRQoL), physical function, anxiety and depression. Methods EpiReumaPt is a national health survey with a three-stage approach. First, 10 661 adult participants were randomly selected. Trained interviewers undertook structured face-to-face questionnaires that included screening for RMDs and assessments of health-related quality of life, physical function, anxiety and depression. Second, positive screenings for ≥1 RMD plus 20% negative screenings were invited to be evaluated by a rheumatologist. Finally, three rheumatologists revised all the information and confirmed the diagnoses according to validated criteria. Estimates were computed as weighted proportions, taking the sampling design into account. Results The disease-specific prevalence rates (and 95% CIs) of RMDs in the adult Portuguese population were: low back pain, 26.4% (23.3% to 29.5%); periarticular disease, 15.8% (13.5% to 18.0%); knee osteoarthritis (OA), 12.4% (11.0% to 13.8%); osteoporosis, 10.2% (9.0% to 11.3%); hand OA, 8.7% (7.5% to 9.9%); hip OA, 2.9% (2.3% to 3.6%); fibromyalgia, 1.7% (1.1% to 2.1%); spondyloarthritis, 1.6% (1.2% to 2.1%); gout, 1.3% (1.0% to 1.6%); rheumatoid arthritis, 0.7% (0.5% to 0.9%); systemic lupus erythaematosus, 0.1% (0.1% to 0.2%) and polymyalgia rheumatica, 0.1% (0.0% to 0.2%). After multivariable adjustment, participants with RMDs had significantly lower EQ5D scores (β=−0.09; p<0.001) and higher HAQ scores (β=0.13; p<0.001) than participants without RMDs. RMDs were also significantly associated with the presence of anxiety symptoms (OR=3.5; p=0.006). Conclusions RMDs are highly prevalent in Portugal and are associated not only with significant physical function and mental health impairment but also with poor HRQoL, leading to more health resource consumption. The EpiReumaPt study emphasises the burden of RMDs in Portugal and the need to increase RMD awareness, being a strong argument to encourage policymakers to increase the amount of resources allocated to the treatment of rheumatic patients.

Evaluating the performance and intrusiveness of virtual machines for desktop grid computing

We experimentally evaluate the performance overhead of the virtual environments VMware Player, QEMU, VirtualPC and VirtualBox on a dual-core machine. Firstly, we assess the performance of a Linux guest OS running on a virtual machine by separately benchmarking the CPU, file I/O and the network bandwidth. These values are compared to the performance achieved when applications are run on a Linux OS directly over the physical machine. Secondly, we measure the impact that a virtual machine running a volunteer @home project worker causes on a host OS. Results show that performance attainable on virtual machines depends simultaneously on the virtual machine software and on the application type, with CPU-bound applications much less impacted than IO-bound ones. Additionally, the performance impact on the host OS caused by a virtual machine using all the virtual CPU, ranges from 10% to 35%, depending on the virtual environment.