Particle methods parallel implementations by GP-GPU strategies

This paper outlines the problems found in the parallelization of SPH (Smoothed Particle Hydrodynamics) algorithms using Graphics Processing Units. Different results of some parallel GPU implementations in terms of the speed-up and the scalability compared to the CPU sequential codes are shown. The most problematic stage in the GPU-SPH algorithms is the one responsible for locating neighboring particles and building the vectors where this information is stored, since these specific algorithms raise many dificulties for a data-level parallelization. Because of the fact that the neighbor location using linked lists does not show enough data-level parallelism, two new approaches have been pro- posed to minimize bank conflicts in the writing and subsequent reading of the neighbor lists. The first strategy proposes an efficient coordination between CPU-GPU, using GPU algorithms for those stages that allow a straight forward parallelization, and sequential CPU algorithms for those instructions that involve some kind of vector reduction. This coordination provides a relatively orderly reading of the neighbor lists in the interactions stage, achieving a speed-up factor of x47 in this stage. However, since the construction of the neighbor lists is quite expensive, it is achieved an overall speed-up of x41. The second strategy seeks to maximize the use of the GPU in the neighbor's location process by executing a specific vector sorting algorithm that allows some data-level parallelism. Al- though this strategy has succeeded in improving the speed-up on the stage of neighboring location, the global speed-up on the interactions stage falls, due to inefficient reading of the neighbor vectors. Some changes to these strategies are proposed, aimed at maximizing the computational load of the GPU and using the GPU texture-units, in order to reach the maximum speed-up for such codes. Different practical applications have been added to the mentioned GPU codes. First, the classical dam-break problem is studied. Second, the wave impact of the sloshing fluid contained in LNG vessel tanks is also simulated as a practical example of particle methods

Applicability of the GP Device to the Circle of Willis Arteries by Using a Mathematical Model

According to the World Health Organization, 15 million people suffer stroke worldwide each year, of these, 5 million die and 5 million are permanently disabled. Stroke is therefore a major cause of mortality world-wide. The majority of strokes are caused by a blood clot that occludes an artery in the brain, and although thrombolytic agents such as Alteplase are used to dissolve clots that arise in the arteries of the brain, there are limitations on the use of these thrombolytic agents. However over the past decade, other methods of treatment have been developed which include Thrombectomy Devices e.g. the 'GP' Thrombus Aspiration Device ('GP' TAD). Such devices may be used as an alternative to thrombolytics or in conjunction with them to extract blood clots in arteries such as the middle cerebral artery of the midbrain brain, and the posterior inferior cerebellar artery (PICA) of the posterior aspect of the brain. In this paper, we mathematically model the removal of blood clots using the 'GP' TAD from selected arteries of the brain where blood clots may arise taking into account factors such as the resistances, compliances and inertances effects. Such mathematical modelling may have potential uses in predicting the pressures necessary to extract blood clots of given lengths, and masses from arteries in the Circle of Willis - posterior circulation of the brain

Connexin-43 prevents hematopoietic stem cell senescence through transfer of reactive oxygen species to bone marrow stromal cells

Hematopoietic stem cell (HSC) aging has become a concern in chemotherapy of older patients. Humoral and paracrine signals from the bone marrow (BM) hematopoietic microenvironment (HM) control HSC activity during regenerative hematopoiesis. Connexin-43 (Cx43), a connexin constituent of gap junctions (GJs) is expressed in HSCs, down-regulated during differentiation, and postulated to be a self-renewal gene. Our studies, however, reveal that hematopoietic-specific Cx43 deficiency does not result in significant long-term competitive repopulation deficiency. Instead, hematopoietic Cx43 (H-Cx43) deficiency delays hematopoietic recovery after myeloablation with 5-fluorouracil (5-FU). 5-FU-treated H-Cx43-deficient HSC and progenitors (HSC/P) cells display decreased survival and fail to enter the cell cycle to proliferate. Cell cycle quiescence is associated with down-regulation of cyclin D1, up-regulation of the cyclin-dependent kinase inhibitors, p21cip1. and p16INK4a, and Forkhead transcriptional factor 1 (Foxo1), and activation of p38 mitogen-activated protein kinase (MAPK), indicating that H-Cx43-deficient HSCs are prone to senescence. The mechanism of increased senescence in H-Cx43-deficient HSC/P cells depends on their inability to transfer reactive oxygen species (ROS) to the HM, leading to accumulation of ROS within HSCs. In vivo antioxidant administration prevents the defective hematopoietic regeneration, as well as exogenous expression of Cx43 in HSC/P cells. Furthermore, ROS transfer from HSC/P cells to BM stromal cells is also rescued by reexpression of Cx43 in HSC/P. Finally, the deficiency of Cx43 in the HM phenocopies the hematopoietic defect in vivo. These results indicate that Cx43 exerts a protective role and regulates the HSC/P ROS content through ROS transfer to the HM, resulting in HSC protection during stress hematopoietic regeneration.

Connexin-43 in the osteogenic BM niche regulates its cellular composition and the bidirectional traffic of hematopoietic stem cells and progenitors

Connexin-43 (Cx43), a gap junction protein involved in control of cell proliferation, differentiation and migration, has been suggested to have a role in hematopoiesis. Cx43 is highly expressed in osteoblasts and osteogenic progenitors (OB/P). To elucidate the biologic function of Cx43 in the hematopoietic microenvironment (HM) and its influence in hematopoietic stem cell (HSC) activity, we studied the hematopoietic function in an in vivo model of constitutive deficiency of Cx43 in OB/P. The deficiency of Cx43 in OB/P cells does not impair the steady state hematopoiesis, but disrupts the directional trafficking of HSC/progenitors (Ps) between the bone marrow (BM) and peripheral blood (PB). OB/P Cx43 is a crucial positive regulator of transstromal migration and homing of both HSCs and progenitors in an irradiated microenvironment. However, OB/P Cx43 deficiency in nonmyeloablated animals does not result in a homing defect but induces increased endosteal lodging and decreased mobilization of HSC/Ps associated with proliferation and expansion of Cxcl12-secreting mesenchymal/osteolineage cells in the BM HM in vivo. Cx43 controls the cellular content of the BM osteogenic microenvironment and is required for homing of HSC/Ps in myeloablated animals

Study of a new 4×3 beam forming network for triangular arrays of three radiating elements

An innovative dissipative multi-beam network for triangular arrays of three radiating elements is proposed. This novel network provides three orthogonal beams in θ0 elevation angle and a fourth one in the broadside steering direction. The network is composed of 90º hybrid couplers and fixed phase shifters. In this paper, a relation between network components, radiating element distance and beam steering directions will be shown. Application of the proposed dissipative network to the triangular cells of three radiating elements that integrate the intelligent antenna GEODA will be exhibited. This system works at 1.7 GHz, it has a 60º single radiating element beamwidth and a distance between array elements of 0.57λ. Both beam patterns, theoretical and simulated, obtained with the network will be depicted. Moreover, the whole system, dissipative network built with GEODA cell array, has been measured in the anechoic chamber of the Radiation Group of Technical University of Madrid, demonstrating expected performance

Potenciación de la respuesta productiva en rumiantes mediante el empleo de proteínas protegidas

El principal objetivo de esta tesis fue incrementar la eficiencia proteica en las dietas de rumiantes mediante el uso de proteínas protegidas (harina de girasol y guisante de primavera), así como mejorar la predicción de los aportes de proteína microbiana. Una partida de harinas comerciales de girasol (HG) y de guisante de primavera (GP) fueron tratadas con soluciones 4 N de ácido málico (268,2 g/L) o ácido ortofosfórico (130,6 g/L). Para cada harina, ácido y día de tratamiento, dos fracciones de 12,5 kg fueron pulverizadas sucesivamente en una hormigonera con la solución de ácido correspondiente mediante un pulverizador de campo. Las dos fracciones fueron mezcladas posteriormente y se dejaron reposar durante 1 h a temperatura ambiente. La mezcla fue luego secada en una estufa de aire forzado a 120 ºC durante 1 h. La estufa fue apagada inmediatamente después y el material tratado se mantuvo dentro de ésta hasta la mañana siguiente. El material fue removido durante el proceso de secado cada 30 min durante las primeras 2 h y cada 60 min durante las 5 h posteriores. Este proceso se repitió hasta conseguir las cantidades de harinas tratadas necesarias en los distintos ensayos. En el primer experimento (capitulo 3) se llevaron a cabo estudios de digestión ruminal e intestinal para evaluar los efectos de la aplicación de las soluciones ácidas indicadas y calor a fin de proteger las proteínas de HG y GP contra la degradación ruminal. Estos estudios se realizaron con tres corderos canulados en el rumen y en el duodeno. El estudio de digestión ruminal fue realizado en tres periodos experimentales en los que los corderos fueron alimentados sucesivamente con tres dietas isoproteicas que incluían HG y GP, sin tratar o tratadas con ácidos málico u ortofosfórico. Cada periodo experimental de 21 días incluyó sucesivamente: 10 días de adaptación a las dietas, un estudio del tránsito ruminal de las partículas de HG y GP (días 11 a 14), y la incubación de las muestras de ambos alimentos en bolsas de nailon (días 15–21). Las harinas incubadas en cada periodo experimental correspondieron a las que fueron incluidas en las dietas. Las bacterias ruminales fueron marcadas desde el día 11 hasta el día 21 del periodo experimental mediante infusión intra-ruminal continua con una fuente de 15N. Tras finalizar las incubaciones in situ el día 21 el rumen fue vaciado en cada periodo para aislar las bacterias asociadas a la fase sólida y liquida del rumen. El estudio de digestión intestinal fue realizado veinte días después del final del estudio ruminal a fin de eliminar el enriquecimiento en 15N de la digesta. En este estudio se incubaron muestras compuestas obtenidas mediante la combinación de los diferentes residuos no degradados en el rumen de forma que fuesen representativas de la composición química de la fracción no degradada en el rumen (RU). En esta fase los corderos fueron alimentados con la dieta sin tratar para determinar la digestibilidad de las harinas tanto tratadas como sin tratar mediante la técnica de las bolsas móviles. Además, las proteínas contenidas en las harinas tratadas y sin tratar, así como en las muestras correspondientes a los residuos a 0 h, las muestras compuestas anteriormente indicadas y las muestras no digeridas intestinalmente fueron extraídas y sometidas a electroforesis para determinar el sitio de digestión de las diferentes fracciones proteicas. Las estimaciones de la RU y la digestibilidad intestinal de la materia seca, la materia orgánica (solamente para RU), la proteína bruta (PB) y el almidón (solamente en GP) fueron obtenidos considerando la contaminación microbiana y las tasas de conminución y salida de partículas. Las estimaciones de RU y de la digestibilidad intestinal disminuyeron en todas las fracciones evaluadas de ambos alimentos al corregir por la contaminación microbiana acaecida en el rumen. Todas las estimaciones de RU aumentaron con los tratamientos de protección, incrementándose también la digestibilidad intestinal de la materia seca en la HG. Los bajos valores de la digestibilidad de la proteína de GP tratado y sin tratar sugieren la presencia de algún factor antitripsico no termolábil es esta harina. Los tratamientos de protección incrementaron consistentemente la fracción de materia seca y PB digerida intestinalmente en los dos alimentos, mientras que la fracción de almidón en la muestra de GP solamente aumentó numéricamente (60,5% de media). Sin embargo, los tratamientos también redujeron la fermentación de la materia orgánica, lo cual podría disminuir la síntesis de proteína microbiana. Los estudios de electroforesis muestran la práctica desaparición de la albumina por la degradación ruminal en ambos alimentos, así como que los cambios en otras proteínas de la muestra RU fueron más pronunciados en GP que en HG. La composición de las bacterias asociadas con las fases de digesta ruminal sólida (BAS) y líquida (BAL) fue estudiada para revisar la precisión de un sistema de predicción previo que determinaba la infravaloración del aporte de nutrientes correspondiente a las BAS cuando de usa 15N como marcador y las BAL como referencia microbiana (capitulo 4). Al comparar con BAS, BAL mostraron menores contenidos en materia orgánica, polisacáridos de glucosa y lípidos totales y un mayor contenido en PB, así como un mayor enriquecimiento en 15N. Los datos obtenidos en el estudio actual se ajustan bien a la ecuación previa que predice el enriquecimiento en 15N de las BAS a partir del mismo valor en BAL. Esta nueva ecuación permite establecer que se produce una infravaloración de un 22% en el aporte de PB al animal a partir de las BAS sintetizadas si las BAL son usadas como muestras de referencia. Una segunda relación calculada utilizando los valores medios por dieta expuestos en numerosos trabajos encontrados en la literatura confirma la magnitud de este error. Esta infravaloración asociada al uso de BAL como referencia fue mayor para el aporte de glucosa (43,1%) y todavía mayor para el aporte de lípidos (59,9%), como consecuencia de los menores contenidos de ambas fracciones en BAL frente a SAB. Estos errores deberían ser considerados para obtener mayor precisión en la estimación del aporte de nutrientes microbianos y mejorar la nutrición de los rumiantes. En el experimento 2 se realizó un estudio de producción (capitulo 5) para evaluar los efectos del tratamiento de las harinas HG y GP con soluciones de ácido málico o ácido ortofosfórico sobre el crecimiento, el consumo de concentrado y el rendimiento y engrasamiento de las canales de corderos de engorde. Noventa corderos machos de cruce entrefino procedentes de tres granjas comerciales (peso inicial medio = 14,6, 15,3 y 13,3 kg, respectivamente) fueron asignados aleatoriamente a cinco dietas con diferentes niveles de proteína y diferentes tratamientos con ácidos y engordados hasta un peso medio al sacrificio de 25 kg. Las fuentes de proteína en el pienso control (C; PB=18,0%) fueron harina de soja, HG y GP sin tratar. En tres de los piensos experimentales, las harinas tratadas con ácido ortofosfórico sustituyeron a las de HG y GP sin tratar (Control Ortofosfórico, PC; PB=18,0% sobre materia seca), sustituyéndose, además, la harina de soja parcialmente (Sustitución Media Ortofosfórico, MSP; PB=16,7%) o totalmente (Sustitución Total Ortofosfórico, TSP; PB=15,6%). Finalmente, en uno de los piensos el ácido ortofosfórico fue reemplazo por acido málico para proteger ambas harinas (Sustitución Media Málico, MSM; PB= 16,7%). La paja de trigo (fuente de forraje) y el concentrado fueron ofrecidos ad libitum. Dieciocho corderos fueron distribuidos en seis cubículos con tres animales para cada dieta. Los datos fueron analizados según un análisis factorial considerando el peso inicial como covariable y la granja de procedencia como bloque. Los datos de consumo de concentrado y eficiencia de conversión fueron analizados usando el cubículo como unidad experimental, mientras que los datos sobre ganancia media diaria, rendimiento a la canal, grasa dorsal y grasa pélvico renal fueron analizados usando el cordero como unidad experimental. No se encontró ningún efecto asociado con el nivel de PB sobre ninguna variable estudiada. Esto sugiere que usando proteínas protegidas es posible utilizar concentrados con 15,6% de PB (sobre materia seca) disminuyendo así la cantidad de concentrados de proteína vegetal a incluir en los piensos y la calidad de los concentrados proteicos. Los corderos alimentados con la dieta MSM tuvieron mayores ganancias medias diarias (15,2%; P= 0,042), y mejores rendimiento a la canal en caliente (1,3 unidades porcentuales; P= 0,037) que los corderos alimentados con el concentrado MSP. Esto podría ser explicado por los efectos benéficos ruminales del malato o por el mayor efecto de protección conseguido con el ácido málico. ABSTRACT The main objective of this thesis project was to increase the protein efficiency in ruminant diets by using protected protein (sunflower meal and spring pea), and improving the prediction of microbial protein supply. Commercial sunflower meal (SFM) and spring pea (SP) were treated with 4 N solutions (200 mL/kg) of malic acid (268.2 g/L) or orthophosphoric acid (130.6 g/L). Daily, two fractions of 12.5 kg of one of these meals were successively sprayed with the tested acid solution in a concrete mixer using a sprayer. Both fractions were then mixed and allowed to rest for 1 h at room temperature. The blend was then dried in a forced air oven at 120 ºC for 1 h. Then the oven was turned off and the treated material was left in the oven overnight. During the drying process, the material was stirred every 30 min during the first 2 h and then every 60 min for the subsequent 5 h. This process was repeated until the amounts of treated flour needed for the different trials performed. In the first experiment (chapter 3), ruminal and intestinal digestion trials were conducted to study the effects of the application of these acid solutions and heat to protect proteins of SFM and SP against ruminal degradation using three wethers fitted with rumen and duodenum cannulae. The ruminal digestion study was carried out in three experimental periods in which the wethers were successively fed three isoproteic diets including SFM and SP, untreated or treated with malic or orthophosphoric acids. The experimental periods of 21 days included successively: 10 days of diet adaptation, SFM and SP particle ruminal transit study (days 11–14) and ruminal nylon-bag incubations (days 15–21). The meals incubated in each experimental period were those corresponding to the associated diet. Rumen bacteria were labelled from days 11 to 21 by continuous intra-ruminal infusion of a 15N source and the rumen was emptied at the end of in situ incubations in each period to isolate solid adherent bacteria and liquid associate bacteria. The intestinal digestion trial was conducted twenty days after the end of the ruminal studies to eliminate the 15N enrichment in the digesta. The tested samples were composite samples obtained pooling the different ruminally undegraded residues to be representative of the chemical composition of the ruminally undegraded fraction (RU). Wethers were fed the untreated diet to determine the intestinal digestibility of untreated and treated meals using the mobile nylon bag technique. In addition, protein in untreated and treated meals and their 0 h, composite and intestinally undigested samples were extracted and subjected to electrophoresis to determine the digestion site of the different protein fractions. Estimates of the RU and its intestinal digestibility of dry matter, organic matter (only for RU), crude protein (CP) and starch (only in SP) were obtained considering ruminal microbial contamination and particle comminution and outflow rates. When corrected for the microbial contamination taking place in the rumen, estimates of RU and intestinal digestibility decreased in all tested fractions for both feeds. All RU estimates increased with the protective treatments, whereas intestinal digestibility-dry matter also increased in SFM. Low intestinal digestibility-CP values in untreated and treated samples suggested the presence of non-heat labile antitrypsin factors in SP. Protective treatments of both feeds led to consistent increases in the intestinal digested fraction of dry matter and CP, being only numerically different for SP-starch (60.5% as average). However, treatments also reduced the organic matter fermentation, which may decrease ruminal microbial protein synthesis. Electrophoretic studies showed albumin disappearance in both SFM and SP, whereas changes in other RU proteins were more pronounced in SP than SFM. The chemical composition of bacteria associated with solid (SAB) and liquid (LAB) rumen-digesta phases was studied to examine the accuracy of a previous regression system determining the underevaluation of SAB-nutrient supply using 15N as marker and LAB as microbial reference (chapter 4). Compared with SAB, LAB showed lower contents of organic matter, polysaccharide-glucose and total lipids and the opposite for the CP content and the 15N enrichment. Present data fitted well to the previous relationship predicting the 15N enrichment of SAB from the same value in LAB. This new equation allows establishing an underevaluation in the supply of CP from the synthesized SAB in 22.0% if LAB is used as reference. Another relationship calculated using mean diet values from the literature confirmed the magnitude of this error. This underevaluation was higher for the supply of glucose (43.1%) and still higher for the lipid supply (59.9%) as a consequence of the lower contents of these both fractions in LAB than in SAB. These errors should be considered to obtain more accurate estimates of the microbial nutrient supply and to improve ruminant nutrition. A production study was performed in experiment 2 (chapter 5) to examine the effects of treating SFM and SP meals with orthophosphoric or malic acid solutions on growth performance, concentrate intake, and carcass yield and fatness of growing-fattening lambs. Ninety "Entrefino" cross male lambs from three commercial farms (average initial body weights (BW) = 14.6, 15.3 and 13.3 kg) were randomly assigned to five diets with different acid treatment and protein levels, and fattened to an average slaughter weight of 25 kg. Protein sources in the control concentrate (C; CP=18%) were soybean meal and untreated SFM and SP. In three of the experimental concentrates, orthophosphoric acid-treated meals substituted untreated SFM and SP (Orthophosphoric Control, PC; CP=18% dry matter basis), and soybean meal was partially (Medium Substitution Orthophosphoric, MSP; CP=16.7%) or totally removed (Total Substitution Orthophosphoric, TSP; CP=15.6%). In addition, in one concentrate orthophosphoric acid was replaced by malic acid to protect these meals (Medium Substitution Malic, MSM; CP= 16.7%). Wheat straw (roughage source) and concentrate were offered ad libitum. Eighteen lambs were allocated to six pens of three animals on each diet. Data were analyzed using a factorial analysis with initial body weight BW as covariate and farm of origin as block. Data on concentrate intake and feed conversion efficiency were analyzed using pen as experimental unit, while data on average daily gain, carcass yield, dorsal fat, and kidney-pelvic-fat were analyzed with lamb as experimental unit. No effect associated with the CP level was observed on any parameter. This suggests that with protected proteins it is possible to feed concentrates with 15.6% CP (dry matter basis) reducing the quantity of vegetable protein meals to include in the concentrate as well as the quality of the protein concentrates. Lambs feed MSM had higher average daily gains (15.2%; P= 0.042), and better hot carcass yields (1.3 percentage points; P= 0.037) than lambs feed MSP. This probably can be explained by ruminal malate actions and by greater protection effects obtained with malic acid.