A User-Centric Perspective on Parallel Programming with Focus on OpenMP

The process of developing software that takes advantage of multiple processors is commonly referred to as parallel programming. For various reasons, this process is much harder than the sequential case. For decades, parallel programming has been a problem for a small niche only: engineers working on parallelizing mostly numerical applications in High Performance Computing. This has changed with the advent of multi-core processors in mainstream computer architectures. Parallel programming in our days becomes a problem for a much larger group of developers. The main objective of this thesis was to find ways to make parallel programming easier for them. Different aims were identified in order to reach the objective: research the state of the art of parallel programming today, improve the education of software developers about the topic, and provide programmers with powerful abstractions to make their work easier. To reach these aims, several key steps were taken. To start with, a survey was conducted among parallel programmers to find out about the state of the art. More than 250 people participated, yielding results about the parallel programming systems and languages in use, as well as about common problems with these systems. Furthermore, a study was conducted in university classes on parallel programming. It resulted in a list of frequently made mistakes that were analyzed and used to create a programmers' checklist to avoid them in the future. For programmers' education, an online resource was setup to collect experiences and knowledge in the field of parallel programming - called the Parawiki. Another key step in this direction was the creation of the Thinking Parallel weblog, where more than 50.000 readers to date have read essays on the topic. For the third aim (powerful abstractions), it was decided to concentrate on one parallel programming system: OpenMP. Its ease of use and high level of abstraction were the most important reasons for this decision. Two different research directions were pursued. The first one resulted in a parallel library called AthenaMP. It contains so-called generic components, derived from design patterns for parallel programming. These include functionality to enhance the locks provided by OpenMP, to perform operations on large amounts of data (data-parallel programming), and to enable the implementation of irregular algorithms using task pools. AthenaMP itself serves a triple role: the components are well-documented and can be used directly in programs, it enables developers to study the source code and learn from it, and it is possible for compiler writers to use it as a testing ground for their OpenMP compilers. The second research direction was targeted at changing the OpenMP specification to make the system more powerful. The main contributions here were a proposal to enable thread-cancellation and a proposal to avoid busy waiting. Both were implemented in a research compiler, shown to be useful in example applications, and proposed to the OpenMP Language Committee.

Fliesslochformen und Gewindefurchen in dünne Bleche und Rohre aus Kupfer und Kupferlegierungen

This thesis in Thermal Flow Drilling and Flowtap in thin metal sheet and pipes of copper and copper alloys had as objectives to know the comportment of copper and copper alloys sheet metal during the Thermal Flow Drill processes with normal tools, to know the best Speed and Feed machine data for the best bushing quality, to known the best Speed for Form Tapping processes and to know the best bush long in pure copper pipes for water solar interchange equipment. Thermal Flow Drilling (TFD) and Form Tapping (FT) is one of the research lines of the Institute of Production and Logistics (IPL) at University of Kassel. At December 1995, a work meeting of IPL, Santa Catarina University, Brazil, Buenos Aires University, Argentine, Tarapacá University (UTA), Chile members and the CEO of Flowdrill B.V. was held in Brazil. The group decided that the Manufacturing Laboratory (ML) of UTA would work with pure copper and brass alloys sheet metal and pure copper pipes in order to develop a water interchange solar heater. The Flowdrill BV Company sent tools to Tarapacá University in 1996. In 1999 IPL and the ML carried out an ALECHILE research project promoted by the DAAD and CONICyT in copper sheet metal and copper pipes and sheet metal a-brass alloys. The normal tools are lobed, conical tungsten carbide tool. When rotated at high speed and pressed with high axial force into sheet metal or thin walled tube generated heat softens the metal and allows the drill to feed forward produce a hole and simultaneously form a bushing from the displacement material. In the market exist many features but in this thesis is used short and longs normal tools of TFD. For reach the objectives it was takes as references four qualities of the frayed end bushing, where the best one is the quality class I. It was used pure copper and a-brass alloys sheet metals, with different thickness. It was used different TFD drills diameter for four thread type, from M-5 to M10. Similar to the Aluminium sheet metals studies it was used the predrilling processes with HSS drills around 30% of the TFD diameter (1,5 – 3,0 mm D). In the next step is used only 2,0 mm thick metal sheet, and 9,2 mm TFD diameter for M-10 thread. For the case of pure commercial copper pipes is used for ¾” inch diameter and 12, 8 mm (3/8”) TFD drill for holes for 3/8” pipes and different normal HSS drills for predrilling processes. The chemical sheet metal characteristics were takes as reference for the material behaviour. The Chilean pure copper have 99,35% of Cu and 0,163% of Zinc and the Chilean a-brass alloys have 75,6% of Cu and 24,0% of Zinc. It is used two German a-brass alloys; Nº1 have 61,6% of Cu, 36,03 % of Zinc and 2,2% of Pb and the German a-brass alloys Nº2 have 63,1% of Cu, 36,7% of Zinc and 0% of Pb. The equipments used were a HAAS CNC milling machine centre, a Kistler dynamometer, PC Pentium II, Acquisition card, TESTPOINT and XAct software, 3D measurement machine, micro hardness, universal test machine, and metallographic microscope. During the test is obtained the feed force and momentum curves that shows the material behaviour with TFD processes. In general it is take three phases. It was possible obtain the best machining data for the different sheet of copper and a-brass alloys thick of Chilean materials and bush quality class I. In the case of a-brass alloys, the chemical components and the TFD processes temperature have big influence. The temperature reach to 400º Celsius during the TFD processes and the a-brass alloys have some percents of Zinc the bush quality is class I. But when the a-brass alloys have some percents of Lead who have 200º C melting point is not possible to obtain a bush, because the Lead gasify and the metallographic net broke. During the TFD processes the recrystallization structures occur around the Copper and a-brass alloy bush, who gives more hardness in these zones. When the threads were produce with Form Tapping processes with Flowtap tools, this hardness amount gives a high limit load of the thread when hey are tested in a special support that was developed for it. For eliminated the predrilling processes with normal HSS drills it was developed a compound tool. With this new tool it was possible obtain the best machining data for quality class I bush. For the copper pipes it is made bush without predrilling and the quality class IV was obtained. When it is was used predrilling processes, quality classes I bush were obtained. Then with different HSS drill diameter were obtained different long bush, where were soldering with four types soldering materials between pipes with 3/8” in a big one as ¾”. Those soldering unions were tested by traction test and all the 3/8” pipes broken, and the soldering zone doesn’t have any problem. Finally were developed different solar water interchange heaters and tested. As conclusions, the present Thesis shows that the Thermal Flow Drilling in thinner metal sheets of cooper and cooper alloys needs a predrilling process for frayed end quality class I bushings, similar to thinner sheets of aluminium bushes. The compound tool developed could obtain quality class I bushings and excludes predrilling processes. The bush recrystalization, product of the friction between the tool and the material, the hardness grows and it is advantageous for the Form Tapping. The methodology developed for commercial copper pipes permits to built water solar interchange heaters.

On-the-Fly Maintenance of Series-Parallel Relationships in Fork-Join Multithreaded Programs

A key capability of data-race detectors is to determine whether one thread executes logically in parallel with another or whether the threads must operate in series. This paper provides two algorithms, one serial and one parallel, to maintain series-parallel (SP) relationships "on the fly" for fork-join multithreaded programs. The serial SP-order algorithm runs in O(1) amortized time per operation. In contrast, the previously best algorithm requires a time per operation that is proportional to Tarjan’s functional inverse of Ackermann’s function. SP-order employs an order-maintenance data structure that allows us to implement a more efficient "English-Hebrew" labeling scheme than was used in earlier race detectors, which immediately yields an improved determinacy-race detector. In particular, any fork-join program running in T₁ time on a single processor can be checked on the fly for determinacy races in O(T₁) time. Corresponding improved bounds can also be obtained for more sophisticated data-race detectors, for example, those that use locks. By combining SP-order with Feng and Leiserson’s serial SP-bags algorithm, we obtain a parallel SP-maintenance algorithm, called SP-hybrid. Suppose that a fork-join program has n threads, T₁ work, and a critical-path length of T[subscript ∞]. When executed on P processors, we prove that SP-hybrid runs in O((T₁/P + PT[subscript ∞]) lg n) expected time. To understand this bound, consider that the original program obtains linear speed-up over a 1-processor execution when P = O(T₁/T[subscript ∞]). In contrast, SP-hybrid obtains linear speed-up when P = O(√T₁/T[subscript ∞]), but the work is increased by a factor of O(lg n).

Using petri nets to specify and execute missions for autonomous underwater vehicles

This paper presents the design and implementation of a mission control system (MCS) for an autonomous underwater vehicle (AUV) based on Petri nets. In the proposed approach the Petri nets are used to specify as well as to execute the desired autonomous vehicle mission. The mission is easily described using an imperative programming language called mission control language (MCL) that formally describes the mission execution thread. A mission control language compiler (MCL-C) able to automatically translate the MCL into a Petri net is described and a real-time Petri net player that allows to execute the resulting Petri net onboard an AUV are also presented

Vereda Crucero Puelenje: interfaz periurbana ante a urbanização em Popayán, Colombia

This article is the result of a work of research conducted on the peri-urban interface of Popayan to detect its potential, restrictions and pathologies regarding the urbanization process. The Crucero Puelenje village, taken as a case study, is located in the south of the town center and it is recognized as one of the agricultural pantries of the city.The territory reading was focused on the macro-landscape as a result of the relationship between social groups and their physical and natural environment. Hence, this research process articulated technical and participatory methodologies that allowed determining the socio-cultural, environmental, physical-spatial, economic and political-administrative particularities of the territory with respect to urbanization and local development. Thus, the participatory component with the Crucero Puelenje community, both in the participatory workshops and in the census, gave feedback to the technical component considered as the guiding thread.This research found knowledge applicable to spatial planning and regional development from the study of the socio-environmental and territorial transformations impact caused by the urban sprawl of Popayán, and at the same time, it offered the Crucero Puelenje community the possibility of interpreting its territory and gaining insights about its planning.

IDELabRoute: librería para la gestión de grafos escalable

Las redes son un importante elemento topológico que tiene poco soporte en el software libre. Hay redes que cuentan con millones de nodos, lo que conlleva la necesidad de manejarlas de forma cuidadosa para optimizar los recursos. Consultando el estado del arte, hemos concluido que existe cierta cantidad de librerías de código abierto disponible, que generalmente emplean un modelo de gestión de los grafos que genera una estructura mallada de objetos en memoria precisando grandes cantidades de memoria y tiempos de puesta en marcha elevados. Estas carencias adquieren especial relevancia cuando se trata de manejar grandes redes. Además las librerías analizadas no suelen ser aptas para procesado multihilo por lo que no pueden usarse en entornos de servidores. Para estos casos hemos puesto en marcha el proyecto IDELabRoute la solución propuesta consiste en una librería genérica de análisis de redes “thread-safe” con gestión dinámica de memoria; para lo cual, se usa una arquitectura modular con gestores de memoria intercambiables, que desde distintas fuentes de almacenamiento persistente (i.e. bases de datos o sistemas de ficheros), maneja grafos de forma dinámica atendiendo a criterios espaciales y/o topológicos. Se trata de una solución de compromiso, puesto que el precio a pagar por la reducción del tamaño de los objetos en memoria es un incremento en el tiempo de respuesta, debido a la gestión de memorias con diversos tiempos de respuesta. Se trata, por tanto, de un sistema de gestión de grafos dinámico que permite manejar grandes modelos de redes de forma escalable, por lo que puede resultar adecuado en entornos con pocos recursos en relación al tamaño total de la red. El primer objetivo práctico del proyecto es proporcionar a la comunidad del GIS libre un servicio WPS para el cálculo de rutas

Alfredo Pareja: ensayista y crítico (Del Historiador y el Ensayista)

La autora revisa la ensayística de Alfredo Pareja sobre temas de arte colonial, cultura y literatura. Da cuenta de la evolución de su noción de lo artístico desde 1933, cuando Pareja marca distancia con la generación anterior y con el arielismo vigente, al hacer suya la defensa del realismo social, más adelante, sin embargo, matizaría estos juicios en otros ensayos. Su postura crítica, en términos generales, no defiende la originalidad ni la forma del texto por sí mismas, sino el valor del contenido, el humanismo empieza a ser el hilo conductor de su narrativa e iluminador de su estética. Por otro lado, siente la necesidad de resignificar lo local en su contacto con lo universal, reconociendo el afán de contemporaneidad de su generación, su defensa de lo mestizo está presente en sus reflexiones sobre el barroquismo latinoamericano. Finalmente, resulta interesante que Pareja aproxime dos grandes momentos del arte ecuatoriano: la pintura y escultura colonial de la llamada escuela de Quito, y la propuesta estética de su generación.

El cinismo idealista de Alejandro Carrión (Dossier: Narradores ecuatorianos de la década de 1950)

Este ensayo analiza aspectos de la dilatada labor como periodista, crítico literario, narrador de ficción y, sobre todo, la producción poética del ecuatoriano Alejandro Carrión. Señala que su narrativa, desde muy temprano, evidencia el tono satírico que no abandonaría jamás, tanto como su «vocación festiva de la sedición». Destaca su labor como periodista, principalmente con el seudónimo de Juan sin cielo. Menciona algunos aportes de Carrión en tanto crítico literario, recogidos principalmente en la revista Letras del Ecuador. El hilo conductor del ensayo –a ratos nostálgica semblanza– se nutre de una perspectiva inevitable: ciertos rasgos de su personalidad (su experiencia en «romances y serpientes», la vocación de incisivo y mordaz humorista, su «naturaleza confrontacional» y aguda inteligencia), además de sus dotes de narrador satírico y de poeta.

Role of polyhydroxybutyrate and glycogen as carbon storage compounds in pea and bean bacteroids

Rhizobium leguminosarum synthesizes polyhydroxybutyrate and glycogen as its main carbon storage compounds. To examine the role of these compounds in bacteroid development and in symbiotic efficiency, single and double mutants of R. legumosarum bv. viciae were made which lack polyhydroxybutyrate synthase (phaC), glycogen synthase (glgA), or both. For comparison, a single phaC mutant also was isolated in a bean-nodulating strain of R. leguminosarum bv. phaseoli. In one large glasshouse trial, the growth of pea plants inoculated with the R. leguminosarum bv. viciae phaC mutant were significantly reduced compared with wild-type-inoculated plants. However, in subsequent glasshouse and growth-room studies, the growth of pea plants inoculated with the mutant were similar to wildtype-inoculated plants. Bean plants were unaffected by the loss of polyhydroxybutyrate biosynthesis in bacteroids. Pea plants nodulated by a glycogen synthase mutants or the glgA/phaC double mutant, grew as well as the wild type in growth-room experiments. Light and electron micrographs revealed that pea nodules infected with the glgA mutant accumulated large amounts of starch in the II/III interzone. This suggests that glycogen may be the dominant carbon storage compound in pea bacteroids. Polyhydroxybutyrate was present in bacteria in the infection thread of pea plants but was broken down during bacteroid formation. In nodules infected with a phaC mutant of R. leguminosarum bv. viciae, there was a drop in the amount of starch in the II/III interzone, where bacteroids form. Therefore, we propose a carbon burst hypothesis for bacteroid formation, where polyhydroxybutyrate accumulated by bacteria is degraded to fuel bacteroid differentiation.

Role of polyhydroxybutyrate and glycogen as carbon storage compounds in pea and bean bacteroids

Rhizobium leguminosarum synthesizes polyhydroxybutyrate and glycogen as its main carbon storage compounds. To examine the role of these compounds in bacteroid development and in symbiotic efficiency, single and double mutants of R. legumosarum bv. viciae were made which lack polyhydroxybutyrate synthase (phaC), glycogen synthase (glgA), or both. For comparison, a single phaC mutant also was isolated in a bean-nodulating strain of R. leguminosarum bv. phaseoli. In one large glasshouse trial, the growth of pea plants inoculated with the R. leguminosarum bv. viciae phaC mutant were significantly reduced compared with wild-type-inoculated plants. However, in subsequent glasshouse and growth-room studies, the growth of pea plants inoculated with the mutant were similar to wildtype-inoculated plants. Bean plants were unaffected by the loss of polyhydroxybutyrate biosynthesis in bacteroids. Pea plants nodulated by a glycogen synthase mutants or the glgA/phaC double mutant, grew as well as the wild type in growth-room experiments. Light and electron micrographs revealed that pea nodules infected with the glgA mutant accumulated large amounts of starch in the II/III interzone. This suggests that glycogen may be the dominant carbon storage compound in pea bacteroids. Polyhydroxybutyrate was present in bacteria in the infection thread of pea plants but was broken down during bacteroid formation. In nodules infected with a phaC mutant of R. leguminosarum bv. viciae, there was a drop in the amount of starch in the II/III interzone, where bacteroids form. Therefore, we propose a carbon burst hypothesis for bacteroid formation, where polyhydroxybutyrate accumulated by bacteria is degraded to fuel bacteroid differentiation.

Metabolism of Rhizobium bacteroids

Nitrogen fixation within legume nodules results from a complex metabolic exchange between bacteria of the family Rhizobiaciae and the plant host. Carbon is supplied to the differentiated bacterial cells, termed bacteroids, in the form of dicarboxylic acids to fuel nitrogen fixation. In exchange, fixed nitrogen is transferred to the plant. Both the bacteroid and the plant-derived peribacteroid membrane tightly regulate the exchange of metabolites. In the bacteroid oxidation of dicarboxylic acids via the TCA cycle occurs in an oxygen-limited environment. This restricts the TCA cycle at key points, such as the 2-oxoglutarate dehydrogenase complex, and requires that inputs of carbon and reductant are balanced with outputs from the TCA cycle. This may be achieved by metabolism through accessory pathways that can remove intermediates, reductant, or ATP from the cycle. These include synthesis of the carbon polymers PHB and glycogen and bypass pathways such as the recently identified 2-oxoglutarate decarboxylase reaction in soybean bacteroids. Recent labeling data have shown that bacteroids synthesize and secrete amino acids, which has led to controversy over the role of amino acids in nodule metabolism. Here we review bacteroid carbon metabolism in detail, evaluate the labeling studies that relate to amino acid metabolism by bacteroids, and place the work in context with the genome sequences of Mesorhizobium loti and Sinorhizobium meliloti. We also consider a wider range of metabolic pathways that are probably of great importance to rhizobia in the rhizosphere, during nodule initiation, infection thread development, and bacteroid development.

The threading of [(CuCN)(2)(mu-4,4 '-bpy)] sheets by CuCN chains

Yellow (CuCN)(2)[(CuCN)(2)(mu-4,4'-bpy)], formed in the hydrothermal reaction of CuCN with 4,4'-bipyridine at 453 K, contains two types of infinite CuCN chains. One set of CuCN chains is linked by 4,4'-bpy ligands to form almost flat sheets of composition [(CuCN)(2)(mu-4,4'-bpy)]. Holes in these sheets are aligned to allow pairs of approximately linear, infinite -(CuCN)- chains to thread through them. The closest interatomic approach between copper atoms in the threading chains and host sheets (similar to2.74 Angstrom) does not appear to represent a significant covalent bond as it leads to only a small distortion of the -(CuCN)- chains from linearity The relationship of this material to the previously determined structures of the host [(CuCN)(2)(mu-4,4'-bpy)] sheets and (CuCN)(3)[(CuCN)(2)(mu-4,4'-bPY)](2), in which these sheets are threaded by single -(CuCN)- chains, is discussed.

Infrared study of intercomponent interactions in a switchable hydrogen-bonded rotaxane

The macrocycle in rotaxane 1 is preferentially hydrogen bonded to the succinamide station in the neutral form, but can be moved to the naphthalimide station by one-electron reduction of the latter. The hydrogen bonding between the amide NH groups of the macrocycle and the C=O groups in the binding stations in the thread was studied with IR spectroscopy in different solvents in both states. In addition, the solvent effect on the vibrational frequencies was analyzed; a correlation with the solvent acceptor number (AN) was observed. The conformational switching upon reduction could be detected by monitoring the hydrogen-bond-induced shifts of the v(CO) frequencies of the C=O groups of the succinamide and the reduced naphthalimide stations. The macrocycle was found to shield the encapsulated station from the solvent: wavenumbers of v(CO) bands of the C=O groups residing inside the macrocycle cavity remain unaffected by the solvent polarity.

A comparative study of Java and C performance in two large-scale parallel applications

In the 1990s the Message Passing Interface Forum defined MPI bindings for Fortran, C, and C++. With the success of MPI these relatively conservative languages have continued to dominate in the parallel computing community. There are compelling arguments in favour of more modern languages like Java. These include portability, better runtime error checking, modularity, and multi-threading. But these arguments have not converted many HPC programmers, perhaps due to the scarcity of full-scale scientific Java codes, and the lack of evidence for performance competitive with C or Fortran. This paper tries to redress this situation by porting two scientific applications to Java. Both of these applications are parallelized using our thread-safe Java messaging system—MPJ Express. The first application is the Gadget-2 code, which is a massively parallel structure formation code for cosmological simulations. The second application uses the finite-domain time-difference method for simulations in the area of computational electromagnetics. We evaluate and compare the performance of the Java and C versions of these two scientific applications, and demonstrate that the Java codes can achieve performance comparable with legacy applications written in conventional HPC languages. Copyright © 2009 John Wiley & Sons, Ltd.

Asparagus: a horticultural ballet

Asparagus: A Horticultural Ballet was a live performance and film narrating the rise of capital in the medium of asparagus. The project stemmed from an obscure reference to an art piece of the same name by Waw Pierogi of the band xex. However, the its re-enactment had little to do with the original exploration of the growth and branching patterns of the asparagus plant. Instead, a rigid choreography inspired by Oskar Schlemmer's Triadic Ballet and based on Karl Marx's Capital dictated the movements of six performers in asparagus costumes. Bringing together the organic and the geometric, the ballet investigated the transition from the Fordist assembly line to immaterial labour through a reanimation of modernist abstraction. Being itself the story of abstraction, Capital shows how human relationships are replaced by those between commodities in the joyless grind of endless accumulation. This process results in the transcendent mythical figure of capital, which frames, transfigures and even produces the natural world. Asparagus: A Horticultural Ballet was produced in collaboration with Montreal based band Les Georges Leningrad and commissioned by The Showroom Gallery, London. It was presented live at Conway Hall in London on 6.3.07 and at the Montreal Biennale at SAT on 12.5.07. The performance was accompanied by a film at the Showroom gallery and preceded by a production residency at the Pump House Gallery. The film has subsequently been shown at The Golden Thread Gallery, Bluecoat Liverpool and as part of A-Lot-Ment in Portsmouth. Props from Asparagus: A Horticultural Ballet, were included in The Eagle Document at the Stephen Lawrence Gallery, London.