Análisis de rendimiento de aplicaciones paralelas de memoria compartida: problema N-body

Este trabajo analiza el rendimiento de cuatro nodos de cómputo multiprocesador de memoria compartida para resolver el problema N-body. Se paraleliza el algoritmo serie, y se codifica usando el lenguaje C extendido con OpenMP. El resultado son dos variantes que obedecen a dos criterios de optimización diferentes: minimizar los requisitos de memoria y minimizar el volumen de cómputo. Posteriormente, se realiza un proceso de análisis de las prestaciones del programa sobre los nodos de cómputo. Se modela el rendimiento de las variantes secuenciales y paralelas de la aplicación, y de los nodos de cómputo; se instrumentan y ejecutan los programas para obtener resultados en forma de varias métricas; finalmente se muestran e interpretan los resultados, proporcionando claves que explican ineficiencias y cuellos de botella en el rendimiento y posibles líneas de mejora. La experiencia de este estudio concreto ha permitido esbozar una incipiente metodología de análisis de rendimiento, identificación de problemas y sintonización de algoritmos a nodos de cómputo multiprocesador de memoria compartida.

Periodic orbits of the planar collision restricted 3-body problem

Using the continuation method we prove that the circular and the elliptic symmetric periodic orbits of the planar rotating Kepler problem can be continued into periodic orbits of the planar collision restricted 3–body problem. Additionally, we also continue to this restricted problem the so called “comets orbits”.

Periodic orbits near a heteroclinic loop formed by one dimensional orbit and a 2-dimensional manifold: application to the charged collinear 3-body problem

The paper is devoted to the study of a type of differential systems which appear usually in the study of some Hamiltonian systems with 2 degrees of freedom. We prove the existence of infinitely many periodic orbits on each negative energy level. All these periodic orbits pass near the total collision. Finally we apply these results to study the existence of periodic orbits in the charged collinear 3–body problem.

Aplicaciones Single Program Multiple Data (SPMD) en ambientes distribuidos

Un reto al ejecutar las aplicaciones en un cluster es lograr mejorar las prestaciones utilizando los recursos de manera eficiente, y este reto es mayor al utilizar un ambiente distribuido. Teniendo en cuenta este reto, se proponen un conjunto de reglas para realizar el cómputo en cada uno de los nodos, basado en el análisis de cómputo y comunicaciones de las aplicaciones, se analiza un esquema de mapping de celdas y un método para planificar el orden de ejecución, tomando en consideración la ejecución por prioridad, donde las celdas de fronteras tienen una mayor prioridad con respecto a las celdas internas. En la experimentación se muestra el solapamiento del computo interno con las comunicaciones de las celdas fronteras, obteniendo resultados donde el Speedup aumenta y los niveles de eficiencia se mantienen por encima de un 85%, finalmente se obtiene ganancias de los tiempos de ejecución, concluyendo que si se puede diseñar un esquemas de solapamiento que permita que la ejecución de las aplicaciones SPMD en un cluster se hagan de forma eficiente.

The Electoral Determinants of Collective Remittances: The Mexican 3x1 Program for Migrants

The 3x1 Program for Migrants is a matching grant scheme that seeks to direct the money sent by migrant organizations abroad to the provision of public and social infrastructure, and to productive projects in migrants’ communities of origin. To do so, the municipal, state, and federal administrations match the amount sent by hometown associations by 3 to 1. This opens the door to the political manipulation of the program. We explore the impact of a particular facet of Mexican political life on the operation of the 3x1: its recent democratization and the increasing political competition at the municipal level. Relying on the literature on redistributive politics, we posit that an increasing number of effective parties in elections may have two different effects. On the one hand, the need to cater to more heterogeneous constituencies may increase the provision of public projects. On the other hand, since smaller coalitions are needed to win elections under tighter competition, fewer public and more private (clientelistic) projects could be awarded. Using a unique dataset on the 3x1 Program for Migrants for over 2,400 municipalities in the period 2002 through 2007, we find a lower provision of public goods in electorally competitive jurisdictions. Thus, we remain sceptical about the program success in promoting public goods in politically competitive locations with high migration levels.

Destruction of invariant curves in the restricted circular planar three body problem using comparison of action

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

Effect of vaccination against gonadotrophin-releasing hormone, using Improvac®, on growth performance, body composition, behaviour and acute phase proteins

The objective of this study was to evaluate the effect of vaccination against GnRH on performance traits, pig behaviour and acute phase proteins. A total of 120 pigs (36 non-castrated males, NCM; 36 males to be vaccinated, IM; 24 castratedmales, CM; and 24 females, FE)were controlled in groups of 12 in pens with feeding stations allowing the recording of individual feed intake. The two vaccinations (Improvac®) were applied at a mean age of 77 and 146 days. All pigswere individually weighed every 3 weeks from the mean ages of 74 to 176 days and backfat thickness (BT) and loinmuscle depth (LD) were also recorded ultrasonically. Twelve group-housed pigs for each treatment were video recorded during 2 consecutive days at weeks 9, 11, 20, 21, 23 and 25 of age to score the number of inactive or active pigs in each treatment group by scan sampling. Aggressive behaviour by the feeder and away from the feeder, and mounting behaviour was also scored by focal sampling. Blood samples from 12 NCM, 12 CM and 12 IM were taken to determine the concentration of circulating acute phase protein Pig-MAP atweeks 1, 2, 4, 11, 13, 21 and 25 of age. After slaughter, the number of skin lesions on the left half carcasswas scored. IMpresented overall a higher growth rate and daily feed intake compared to NCM (Pb0.05),whereas their feed conversion ratios did not differ significantly. In comparison with CM, IM presented a better feed conversion ratio (Pb0.05), since their overall dailyweight gaindid not differ significantly, butIM ate less. Final leanmeat percentage of IM and CM was lower compared to that of NCM (Pb0.05). Activity, mounting and aggressive behaviour of NCM was higher than in IM, CM and FE after the second vaccination. Pig-MAP concentrationswere significantly elevated just after surgical castrationand after bothadministrations of the vaccine (Pb0.05), but concentrations subsequently decreased throughout time. Skin lesions of NCM were significantly higher compared to that of IM and FE (Pb0.05). The effects of vaccination were especially remarkable after the second dose, when the higher feed intake and lower activity of IM compared to NCMmight result in higher final body weight and more fat. Results from this study indicate that some welfare aspects such as a reduced aggression and mounting behaviour may be improved by vaccination against GnRH, together with productive benefits like adequate feed conversion ratio and daily weight gain.

Including the workload effect in the parallel program signature

Performance prediction and application behavior modeling have been the subject of exten- sive research that aim to estimate applications performance with an acceptable precision. A novel approach to predict the performance of parallel applications is based in the con- cept of Parallel Application Signatures that consists in extract an application most relevant parts (phases) and the number of times they repeat (weights). Executing these phases in a target machine and multiplying its exeuction time by its weight an estimation of the application total execution time can be made. One of the problems is that the performance of an application depends on the program workload. Every type of workload affects differently how an application performs in a given system and so affects the signature execution time. Since the workloads used in most scientific parallel applications have dimensions and data ranges well known and the behavior of these applications are mostly deterministic, a model of how the programs workload affect its performance can be obtained. We create a new methodology to model how a program’s workload affect the parallel application signature. Using regression analysis we are able to generalize each phase time execution and weight function to predict an application performance in a target system for any type of workload within predefined range. We validate our methodology using a synthetic program, benchmarks applications and well known real scientific applications.