Applying RADIC in Open MPI: the methodology used to implement RADIC over a Message Passing Library

Fault tolerance has become a major issue for computer and software engineers because the occurrence of faults increases the cost of using a parallel computer. RADIC is the fault tolerance architecture for message passing systems which is transparent, decentralized, flexible and scalable. This master thesis presents the methodology used to implement the RADIC architecture over Open MPI, a well-know large-used message passing library. This implementation kept the RADIC architecture characteristics. In order to validate the implementation we have executed a synthetic ping program, besides, to evaluate the implementation performance we have used the NAS Parallel Benchmarks. The results prove that the RADIC architecture performance depends on the communication pattern of the parallel application which is running. Furthermore, our implementation proves that the RADIC architecture could be implemented over an existent message passing library.

Localization of algebras over coloured operads

We give sufficient conditions for homotopical localization functors to preserve algebras over coloured operads in monoidal model categories. Our approach encompasses a number of previous results about preservation of structures under localizations, such as loop spaces or infinite loop spaces, and provides new results of the same kind. For instance, under suitable assumptions, homotopical localizations preserve ring spectra (in the strict sense, not only up to homotopy), modules over ring spectra, and algebras over commutative ring spectra, as well as ring maps, module maps, and algebra maps. It is principally the treatment of module spectra and their maps that led us to the use of coloured operads (also called enriched multicategories) in this context.

Experimental investigation of transient natural convection cooling of high prandtl number fluid with variable viscosity

Report for the scientific sojourn at the James Cook University, Australia, between June to December 2007. Free convection in enclosed spaces is found widely in natural and industrial systems. It is a topic of primary interest because in many systems it provides the largest resistance to the heat transfer in comparison with other heat transfer modes. In such systems the convection is driven by a density gradient within the fluid, which, usually, is produced by a temperature difference between the fluid and surrounding walls. In the oil industry, the oil, which has High Prandtl, usually is stored and transported in large tanks at temperatures high enough to keep its viscosity and, thus the pumping requirements, to a reasonable level. A temperature difference between the fluid and the walls of the container may give rise to the unsteady buoyancy force and hence the unsteady natural convection. In the initial period of cooling the natural convection regime dominates over the conduction contribution. As the oil cools down it typically becomes more viscous and this increase of viscosity inhibits the convection. At this point the oil viscosity becomes very large and unloading of the tank becomes very difficult. For this reason it is of primary interest to be able to predict the cooling rate of the oil. The general objective of this work is to develop and validate a simulation tool able to predict the cooling rates of high Prandtl fluid considering the variable viscosity effects.

Creation of a carrier-grade telco based on a voice over IP backbone

En aquesta memòria l'autor, fent servir un enfoc modern, redissenya i implementa la plataforma que una empresa de telecomunicacions del segle 21 necessita per poder donar serveis de telefonia i comunicacions als seus usuaris i clients. Al llarg d'aquesta exposició es condueix al lector des d'una fase inicial de disseny fins a la implementació i posada en producció del sistema final desenvolupat, centrant-nos en solucionar les necessitats actuals que això implica. Aquesta memòria cubreix el software, hardware i els processos de negoci associats al repte de fer realitat aquest objectiu, i presenta al lector les múltiples tecnologies emprades per aconseguir-ho, fent emfàsi en la convergència actual de xarxes cap al concepte de xarxes IP i basant-se en aquesta tendència i utilitzant aquesta tecnologia de veu sobre IP per donar forma a la plataforma que finalment, de forma pràctica, es posa en producció.

Persistent bundles over a two dimensional compact set

"Vegeu el resum a l'inici del document del fitxer adjunt."

Are the high-growth recovery periods over?

We present evidence about the loss of the so-called "plucking effect", that is, a high-growth phase of the cycle typically observed at the end of recessions. This result matches the popular belief, presented informally by different authors, that the current recession will have permanent effects, or that the current recession will have an L shape versus the old-time recessions that have always had a V shape. Furthermore, we show that the loss of the "plucking effect" can explain part of the Great Moderation. We postulate that these two phenomena may be due to changes in inventory management brought about by improvements in information and communications technologies.

On the density distribution across space: a probabilistic approach

This paper aims at providing a Bayesian parametric framework to tackle the accessibility problem across space in urban theory. Adopting continuous variables in a probabilistic setting we are able to associate with the distribution density to the Kendall's tau index and replicate the general issues related to the role of proximity in a more general context. In addition, by referring to the Beta and Gamma distribution, we are able to introduce a differentiation feature in each spatial unit without incurring in any a-priori definition of territorial units. We are also providing an empirical application of our theoretical setting to study the density distribution of the population across Massachusetts.

A Data mining approach to indirect inference

Consider a model with parameter phi, and an auxiliary model with parameter theta. Let phi be a randomly sampled from a given density over the known parameter space. Monte Carlo methods can be used to draw simulated data and compute the corresponding estimate of theta, say theta_tilde. A large set of tuples (phi, theta_tilde) can be generated in this manner. Nonparametric methods may be use to fit the function E(phi|theta_tilde=a), using these tuples. It is proposed to estimate phi using the fitted E(phi|theta_tilde=theta_hat), where theta_hat is the auxiliary estimate, using the real sample data. This is a consistent and asymptotically normally distributed estimator, under certain assumptions. Monte Carlo results for dynamic panel data and vector autoregressions show that this estimator can have very attractive small sample properties. Confidence intervals can be constructed using the quantiles of the phi for which theta_tilde is close to theta_hat. Such confidence intervals are found to have very accurate coverage.

Aspects of Iwasawa theory over function fields

"Vegeu el resum a l'inici del document del fitxer adjunt."

Sobolev mappings: Lipschitz density is not an isometric invariant of the target

"Vegeu el resum a l'inici del document del fitxer adjunt."

What are the causes of educational inequalities and of their evolution over time in Europe? Evidence from PISA

This paper provides evidence on the sources of differences in inequalities in educational scores in European Union member states, by decomposing them into their determining factors. Using PISA data from the 2000 and 2006 waves, the paper shows that inequalities emerge in all countries and in both period, but decreased in Germany, whilst they increased in France and Italy. Decomposition shows that educational inequalities do not only reflect background related inequality, but especially schools’ characteristics. The findings allow policy makers to target areas that may make a contribution in reducing educational inequalities.

Gaussian estimates for the density of the non-linear stochastic heat equation in any space dimension

In this paper, we establish lower and upper Gaussian bounds for the probability density of the mild solution to the stochastic heat equation with multiplicative noise and in any space dimension. The driving perturbation is a Gaussian noise which is white in time with some spatially homogeneous covariance. These estimates are obtained using tools of the Malliavin calculus. The most challenging part is the lower bound, which is obtained by adapting a general method developed by Kohatsu-Higa to the underlying spatially homogeneous Gaussian setting. Both lower and upper estimates have the same form: a Gaussian density with a variance which is equal to that of the mild solution of the corresponding linear equation with additive noise.

Regional value added in Italy over the long run (1891-2001): linking indirect estimates with official figures, and implications

This paper presents value added estimates for the Italian regions, in benchmark years from 1891 until 1951, which are linked to those from official figures available from 1971 in order to offer a long-term picture. Sources and methodology are documented and discussed, whilst regional activity rates and productivity are also presented and compared. Thus some questions are briefly reconsidered: the origins and extent of the north-south divide, the role of migration and regional policy in shaping the pattern of regional inequality, the importance of social capital, and the positioning of Italy in the international debate on regional convergence, where it stands out for the long run persistence of its disparities.