Search

Multibody systems consist of a set of components connected through some joints, where the movement of the system is determined by those of its components. Their design is computationally demanding, and the group equations formulation facilitates the application of parallelism to reduce the simulation time. A simulator for the kinematic analysis of multibody systems on up-to-date computational nodes (multicore CPU+GPU) is presented. The movement of the components is simulated by repeatedly solving independent linear systems working on sparse matrices. The appropriate selection of the linear algebra library to be used and the degree of parallelism at each level (explicit with OpenMP and implicit with multithread libraries) help obtain important reductions in the simulation time.

Social networks are known to be assortative with respect to many attributes, such as age, weight, wealth, level of education, ethnicity and gender. This can be explained by influences and homophilies. Independently of its origin, this assortativity gives us information about each node given its neighbors. Assortativity can thus be used to improve individual predictions in a broad range of situations, when data are missing or inaccurate. This paper presents a general framework based on probabilistic graphical models to exploit social network structures for improving individual predictions of node attributes. Using this framework, we quantify the assortativity range leading to an accuracy gain in several situations. We finally show how specific characteristics of the network can improve performances further. For instance, the gender assortativity in real-world mobile phone data changes significantly according to some communication attributes. In this case, individual predictions with 75% accuracy are improved by up to 3%.

Missing requirements are known to be among the major causes of software failure. They often result from a natural inclination to conceive over-ideal systems where the software-to-be and its environment always behave as expected. Obstacle analysis is a goal-anchored form of risk analysis whereby exceptional conditions that may obstruct system goals are identified, assessed and resolved to produce complete requirements. Various techniques have been proposed for identifying obstacle conditions systematically. Among these, the formal ones have limited applicability or are costly to automate. This paper describes a tool-supported technique for generating a set of obstacle conditions guaranteed to be complete and consistent with respect to the known domain properties. The approach relies on a novel combination of model checking and learning technologies. Obstacles are iteratively learned from counterexample and witness traces produced by model checking against a goal and converted into positive and negative examples, respectively. A comparative evaluation is provided with respect to published results on the manual derivation of obstacles in a real safety-critical system for which failures have been reported.

Before developing any Service-Oriented Architecture (SOA) based software, effort estimation will be inevitably required for many purposes ranging from budget analysis to cost-benefit balance. Therefore, effort estimation plays a vital role in SOA implementations. Nevertheless, effort estimation for SOA implementation confronts many challenges due to both the complexity of SOA and the diversity of SOA projects. The current approaches based on traditional techniques are normally not adequate for estimating the effort of SOA implementations. Meanwhile, there is a lack of work on effort estimation for SOA implementation within the published literature. To decrease the complexity and difficulty of the problems of effort estimation before building SOA systems, this thesis concentrates on one particular type of SOA projects: service composition based SOA implementations. As enterprises move to having more and more services, and business applications increasingly rely on subscribing services, the major problem in SOA implementations will be service composition and may be less on development of new services. Moreover, it has been identified that the benefits of SOA cannot be fully realized until reaching the level of service composition. Therefore, it is worthwhile to narrow down our concerns from generic SOA implementations to service composition based SOA implementations that do not take into account service migration or new service development. Considering Web service is the de facto representation of service in practice, we can further focus on Web service composition (WSC) based SOA implementations.Considering the limitations to empirically implementing SOA projects by using various WSC approaches, “learning by doing” is nearly an impossible way to collect development data for quantitatively unfolding the research into effort estimation for WSC-based SOA implementations. Fortunately, “learning by analogy” is a feasible alternative to investigate dist

In this paper, we formulate the lp-norm optimization problem as a conic optimization problem, derive its duality properties (weak duality, zero duality gap, and primal attainment) using standard conic duality and show how it can be solved in polynomial time applying the framework of interior-point algorithms based on self-concordant barriers.

International audience, Object-oriented frameworks are often difficult to use. Framework- specific extensions to integrated development environments (IDEs) aim to mitigate the difficulty by offering tools that leverage the knowledge about framework's application programming interfaces (APIs). These tools commonly offer support for code visualization, automatic and interactive code generation, and code validation. Current practices, however, require such extensions to be custom-built manually for each supported framework. In this paper, we propose an approach to building framework-specific IDE extensions based on framework-specific modeling languages (FSMLs). We show how the definitions of different FSMLs can be interpreted in these extensions to provide advanced tool support for different framework APIs that the FSMLs are designed for.

International audience, We investigate the problem of the verification of multiagent systems by means of parallel algorithms. We present algorithms for CTLK, a logic combining branching time temporal logic with epistemic modalities. We report on an implementation of these algorithms and present the experimental results obtained. The results point to a significant speed-up in the verification step.