Aceleración de Algoritmos en Procesadores Multicore, GPGPU y FPGA. Algorithm Acceleration with Multicore, GPGPU and FPGA Processors.

El avance en la potencia de cómputo en nuestros días viene dado por la paralelización del procesamiento, dadas las características que disponen las nuevas arquitecturas de hardware. Utilizar convenientemente este hardware impacta en la aceleración de los algoritmos en ejecución (programas). Sin embargo, convertir de forma adecuada el algoritmo en su forma paralela es complejo, y a su vez, esta forma, es específica para cada tipo de hardware paralelo. En la actualidad los procesadores de uso general más comunes son los multicore, procesadores paralelos, también denominados Symmetric Multi-Processors (SMP). Hoy en día es difícil hallar un procesador para computadoras de escritorio que no tengan algún tipo de paralelismo del caracterizado por los SMP, siendo la tendencia de desarrollo, que cada día nos encontremos con procesadores con mayor numero de cores disponibles. Por otro lado, los dispositivos de procesamiento de video (Graphics Processor Units - GPU), a su vez, han ido desarrollando su potencia de cómputo por medio de disponer de múltiples unidades de procesamiento dentro de su composición electrónica, a tal punto que en la actualidad no es difícil encontrar placas de GPU con capacidad de 200 a 400 hilos de procesamiento paralelo. Estos procesadores son muy veloces y específicos para la tarea que fueron desarrollados, principalmente el procesamiento de video. Sin embargo, como este tipo de procesadores tiene muchos puntos en común con el procesamiento científico, estos dispositivos han ido reorientándose con el nombre de General Processing Graphics Processor Unit (GPGPU). A diferencia de los procesadores SMP señalados anteriormente, las GPGPU no son de propósito general y tienen sus complicaciones para uso general debido al límite en la cantidad de memoria que cada placa puede disponer y al tipo de procesamiento paralelo que debe realizar para poder ser productiva su utilización. Los dispositivos de lógica programable, FPGA, son dispositivos capaces de realizar grandes cantidades de operaciones en paralelo, por lo que pueden ser usados para la implementación de algoritmos específicos, aprovechando el paralelismo que estas ofrecen. Su inconveniente viene derivado de la complejidad para la programación y el testing del algoritmo instanciado en el dispositivo. Ante esta diversidad de procesadores paralelos, el objetivo de nuestro trabajo está enfocado en analizar las características especificas que cada uno de estos tienen, y su impacto en la estructura de los algoritmos para que su utilización pueda obtener rendimientos de procesamiento acordes al número de recursos utilizados y combinarlos de forma tal que su complementación sea benéfica. Específicamente, partiendo desde las características del hardware, determinar las propiedades que el algoritmo paralelo debe tener para poder ser acelerado. Las características de los algoritmos paralelos determinará a su vez cuál de estos nuevos tipos de hardware son los mas adecuados para su instanciación. En particular serán tenidos en cuenta el nivel de dependencia de datos, la necesidad de realizar sincronizaciones durante el procesamiento paralelo, el tamaño de datos a procesar y la complejidad de la programación paralela en cada tipo de hardware. Today´s advances in high-performance computing are driven by parallel processing capabilities of available hardware architectures. These architectures enable the acceleration of algorithms when thes ealgorithms are properly parallelized and exploit the specific processing power of the underneath architecture. Most current processors are targeted for general pruposes and integrate several processor cores on a single chip, resulting in what is known as a Symmetric Multiprocessing (SMP) unit. Nowadays even desktop computers make use of multicore processors. Meanwhile, the industry trend is to increase the number of integrated rocessor cores as technology matures. On the other hand, Graphics Processor Units (GPU), originally designed to handle only video processing, have emerged as interesting alternatives to implement algorithm acceleration. Current available GPUs are able to implement from 200 to 400 threads for parallel processing. Scientific computing can be implemented in these hardware thanks to the programability of new GPUs that have been denoted as General Processing Graphics Processor Units (GPGPU).However, GPGPU offer little memory with respect to that available for general-prupose processors; thus, the implementation of algorithms need to be addressed carefully. Finally, Field Programmable Gate Arrays (FPGA) are programmable devices which can implement hardware logic with low latency, high parallelism and deep pipelines. Thes devices can be used to implement specific algorithms that need to run at very high speeds. However, their programmability is harder that software approaches and debugging is typically time-consuming. In this context where several alternatives for speeding up algorithms are available, our work aims at determining the main features of thes architectures and developing the required know-how to accelerate algorithm execution on them. We look at identifying those algorithms that may fit better on a given architecture as well as compleme

A Novel Algorithm to Quantify Coronary Remodeling Using Inferred Normal Dimensions

Background:Vascular remodeling, the dynamic dimensional change in face of stress, can assume different directions as well as magnitudes in atherosclerotic disease. Classical measurements rely on reference to segments at a distance, risking inappropriate comparison between dislike vessel portions.Objective:to explore a new method for quantifying vessel remodeling, based on the comparison between a given target segment and its inferred normal dimensions.Methods:Geometric parameters and plaque composition were determined in 67 patients using three-vessel intravascular ultrasound with virtual histology (IVUS-VH). Coronary vessel remodeling at cross-section (n = 27.639) and lesion (n = 618) levels was assessed using classical metrics and a novel analytic algorithm based on the fractional vessel remodeling index (FVRI), which quantifies the total change in arterial wall dimensions related to the estimated normal dimension of the vessel. A prediction model was built to estimate the normal dimension of the vessel for calculation of FVRI.Results:According to the new algorithm, “Ectatic” remodeling pattern was least common, “Complete compensatory” remodeling was present in approximately half of the instances, and “Negative” and “Incomplete compensatory” remodeling types were detected in the remaining. Compared to a traditional diagnostic scheme, FVRI-based classification seemed to better discriminate plaque composition by IVUS-VH.Conclusion:Quantitative assessment of coronary remodeling using target segment dimensions offers a promising approach to evaluate the vessel response to plaque growth/regression.

Algorithm engineering for expression dag based number types

Magdeburg, Univ., Fak. für Informatik, Diss., 2015

Water depth selection during foraging and efficiency in prey capture by the egrets Casmerodius albus and Egretta thula (Aves, Ardeidae) in an urban lagoon in Rio de Janeiro State, Brazil

This study aimed to evaluate the water depth selection during foraging, the efficiency in prey capture, and the food items captured by Casmerodius albus (Linnaeus, 1758) and Egretta thula (Molina, 1782). The work was conducted at an urban lagoon, Lagoa Rodrigo de Freitas, Rio de Janeiro. Four transects were made each month (two in the morning and two in the afternoon) for six months. When the birds were detected foraging, the water depth and the types of prey captured were recorded. There was no significant relationship between the foraging efficiencies of the two species. However, they differed in relation to the water depth when foraging, and also in the food items captured. Casmerodius albus captured mainly fishes while Egretta thula captured mainly invertebrates. The results suggest that the differences in water depth when foraging and the food items captured allow a differential use of the food resources available by C. albus and E. thula at Lagoa Rodrigo de Freitas.

Trophic ecology and foraging behavior of Tropidurus hispidus and Tropidurus semitaeniatus (Squamata, Tropiduridae) in a caatinga area of northeastern Brazil

This study aimed to analyze the seasonal variation in diet composition and foraging behavior of Tropidurus hispidus (Spix, 1825) and T. semitaeniatus (Spix, 1825), as well as measurement of the foraging intensity (number of moves, time spent stationary, distance traveled and number of attacks on prey items) in a caatinga patch on the state of Rio Grande do Norte, Brazil. Hymenoptera/Formicidae and Isoptera predominated in the diet of both species during the dry season. Opportunistic predation on lepidopteran larvae, coleopteran larvae and adults, and orthopteran nymphs and adults occurred in the wet season; however, hymenopterans/Formicidae were the most important prey items. The number of food items was similar between lizard species in both seasons; however the overlap for number of prey was smaller in the wet season. Preys ingested by T. hispidus during the wet season were also larger than those consumed by T. semitaeniatus. Seasonal comparisons of foraging intensity between the two species differed, mainly in the wet season, when T. hispidus exhibited less movement and fewer attacks on prey, and more time spent stationary if compared to T. semitaeniatus. Although both lizards are sit-and-wait foragers, T. semitaeniatus is more active than T. hispidus. The diet and foraging behavior of T. hispidus and T. semitaeniatus overlap under limiting conditions during the dry season, and are segregative factors that may contribute to the coexistence of these species in the wet season.

Foraging behavioral of Phylloscartes ventralis (Aves, Tyrannidae) in native and planted forests of southern Brazil

Few studies have related the effects of silviculture practices to the behavior of bird species in the Neotropics. The present study examined the foraging behavior of Phylloscartes ventralis (Temminck, 1824) in a native forest and in silviculture areas of Pinus elliotti and Araucaria angustifolia with different structures and ages. We tested two general hypotheses: (1) areas of commercial forest plantation change the foraging behavior of P. ventralis in relation to native forest, and (2) the foraging behavior of P. ventralis in silviculture areas with understories (complex structures) is different from its behavior in areas without understory. The results showed that P. ventralis changed its foraging behavior depending on the type of forest, and on the presence of an understory in silviculture areas. Main changes involved the height and angle of substrate where the prey was captured. Phylloscartes ventralis showed the same set of attack maneuvers, with more maneuvers type in young Pinus planted without understory. The frequency of use of attack maneuvers was more similar in areas of silviculture with understory and in the native forest. The results highlight the importance of an understory structure and the utilization of native plant species in silviculture practices, to the foraging behavior of native bird species.

A note on parallelizing the parameterized expectations algorithm

The parameterized expectations algorithm (PEA) involves a long simulation and a nonlinear least squares (NLS) fit, both embedded in a loop. Both steps are natural candidates for parallelization. This note shows that parallelization can lead to important speedups for the PEA. I provide example code for a simple model that can serve as a template for parallelization of more interesting models, as well as a download link for an image of a bootable CD that allows creation of a cluster and execution of the example code in minutes, with no need to install any software.

Some characteristics of hyperreactivity to bacterial lipopolysaccharide induced in mice by Trypanosoma cruzi infection

Mice infected with T. cruzi strain, acquire a high level of susceptibility to the effects of bacterial gram-negative LPS. The LD50 of adult female SW mice to LPS from S. typhosa, decreases from 450 to 2,5 mcg 10-12 days after T. cruzi infection. This hyperreactivity to LPS induced by T. cruzi presents all the characteristics of that found in infection caused by many other agents. During the acaute phase of experimental infection with T. cruzi Y strain, mice generally die with a hypovolemic shock very similar to that induced in uninfected animals injected with an adequate dose of bacterial endotoxin. There is evidence for and against the hypothesis that LPS absorbed from the instestinal tract may be involved in the mechanism of death of mice during the acute phase of T. cruzi infection.

Endocytosis-inducer adhesins produced by enteropathogenic serogroups of Escherichia coli participate on bacterial attachment to infant enterocytes

Enteropathogenic E. coli (EPEC) infection of Hep-2 cells preoceeds through bacterial attachment to cell surface and internalization of adhered bacteria. EPEC attachment is a prerequisite for cell infection and is mediated by adhesins that recognize carbohydrate-containing receptors on cell membrane. Such endocytosis-inducer adhesins (EIA) also promote EPEC binding to infant enterocytes, suggesting that EIA may have an important role on EPEC gastroenteritis.

Individual-based modelling of bacterial cultures to study the lag ohase

Els bacteris són la forma dominant de vida del planeta: poden sobreviure en medis molt adversos, i en alguns casos poden generar substàncies que quan les ingerim ens són tòxiques. La seva presència en els aliments fa que la microbiologia predictiva sigui un camp imprescindible en la microbiologia dels aliments per garantir la seguretat alimentària. Un cultiu bacterià pot passar per quatre fases de creixement: latència, exponencial, estacionària i de mort. En aquest treball s’ha avançat en la comprensió dels fenòmens intrínsecs a la fase de latència, que és de gran interès en l’àmbit de la microbiologia predictiva. Aquest estudi, realitzat al llarg de quatre anys, s’ha abordat des de la metodologia Individual-based Modelling (IbM) amb el simulador INDISIM (INDividual DIScrete SIMulation), que ha estat millorat per poder fer-ho. INDISIM ha permès estudiar dues causes de la fase de latència de forma separada, i abordar l’estudi del comportament del cultiu des d’una perspectiva mesoscòpica. S’ha vist que la fase de latència ha de ser estudiada com un procés dinàmic, i no definida per un paràmetre. L’estudi de l’evolució de variables com la distribució de propietats individuals entre la població (per exemple, la distribució de masses) o la velocitat de creixement, han permès distingir dues etapes en la fase de latència, inicial i de transició, i aprofundir en la comprensió del que passa a nivell cel•lular. S’han observat experimentalment amb citometria de flux diversos resultats previstos per les simulacions. La coincidència entre simulacions i experiments no és trivial ni casual: el sistema estudiat és un sistema complex, i per tant la coincidència del comportament al llarg del temps de diversos paràmetres interrelacionats és un aval a la metodologia emprada en les simulacions. Es pot afirmar, doncs, que s’ha verificat experimentalment la bondat de la metodologia INDISIM.

Performance of matrix-assisted laser desorption ionization-time of flight mass spectrometry for identification of bacterial strains routinely isolated in a clinical microbiology laboratory.

Matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) has recently been introduced in diagnostic microbiology laboratories for the identification of bacterial and yeast strains isolated from clinical samples. In the present study, we prospectively compared MALDI-TOF MS to the conventional phenotypic method for the identification of routine isolates. Colonies were analyzed by MALDI-TOF MS either by direct deposition on the target plate or after a formic acid-acetonitrile extraction step if no valid result was initially obtained. Among 1,371 isolates identified by conventional methods, 1,278 (93.2%) were putatively identified to the species level by MALDI-TOF MS and 73 (5.3%) were identified to the genus level, but no reliable identification was obtained for 20 (1.5%). Among the 1,278 isolates identified to the species level by MALDI-TOF MS, 63 (4.9%) discordant results were initially identified. Most discordant results (42/63) were due to systematic database-related taxonomical differences, 14 were explained by poor discrimination of the MALDI-TOF MS spectra obtained, and 7 were due to errors in the initial conventional identification. An extraction step was required to obtain a valid MALDI-TOF MS identification for 25.6% of the 1,278 valid isolates. In conclusion, our results show that MALDI-TOF MS is a fast and reliable technique which has the potential to replace conventional phenotypic identification for most bacterial strains routinely isolated in clinical microbiology laboratories.

CD14 expression on monocytes and TNF alpha production in patients with septic shock, cardiogenic shock or bacterial pneumonia.

OBJECTIVES: In patients with septic shock, circulating monocytes become refractory to stimulation with microbial products. Whether this hyporesponsive state is induced by infection or is related to shock is unknown. To address this question, we measured TNF alpha production by monocytes or by whole blood obtained from healthy volunteers (controls), from patients with septic shock, from patients with severe infection (bacterial pneumonia) without shock, and from patients with cardiogenic shock without infection. MEASUREMENTS: The numbers of circulating monocytes, of CD14+ monocytes, and the expression of monocyte CD14 and the LPS receptor, were assessed by flow cytometry. Monocytes or whole blood were stimulated with lipopolysaccharide endotoxin (LPS), heat-killed Escherichia coli or Staphylococcus aureus, and TNF alpha production was measured by bioassay. RESULTS: The number of circulating monocytes, of CD14+ monocytes, and the monocyte CD14 expression were significantly lower in patients with septic shock than in controls, in patients with bacterial pneumonia or in those with cardiogenic shock (p < 0.001). Monocytes or whole blood of patients with septic shock exhibited a profound deficiency of TNF alpha production in response to all stimuli (p < 0.05 compared to controls). Whole blood of patients with cardiogenic shock also exhibited this defect (p < 0.05 compared to controls), although to a lesser extent, despite normal monocyte counts and normal CD14 expression. CONCLUSIONS: Unlike patients with bacterial pneumonia, patients with septic or cardiogenic shock display profoundly defective TNF alpha production in response to a broad range of infectious stimuli. Thus, down-regulation of cytokine production appears to occur in patients with systemic, but not localised, albeit severe, infections and also in patients with non-infectious circulatory failure. Whilst depletion of monocytes and reduced monocyte CD14 expression are likely to be critical components of the hyporesponsiveness observed in patients with septic shock, other as yet unidentified factors are at work in this group and in patients with cardiogenic shock.

Pullulation bactérienne de l'intestin grêle [Small intestine bacterial overgrowth]

Small intestine bacterial overgrowth (SIBO) is a condition characterised by nutrient malabsorption and excessive bacteria in the small intestine. It typically presents with diarrhea, flatulence and a syndrome of malabsorption (steatorrhea, macrocytic anemia). However, it may be asymptomatic in the eldery. A high index of suspicion is necessary in order to differentiate SIBO from other similar presenting disorders such as coeliac disease, lactose intolerance or the irritable bowel syndrome. A search for predisposing factor is thus necessary. These factors may be anatomical (stenosis, blind loop), or functional (intestinal hypomotility, achlorydria). The hydrogen breath test is the most frequently used diagnostic test although it lacks standardisation. The treatment of SIBO consists of eliminating predisposing factors and broad-spectrum antibiotic therapy.