La tendance actuelle des constructeurs pour le calcul scientifique est à l'utilisation de grappes de machines dont les noeuds comportent un nombre de coeurs toujours plus grand. Le modèle basé uniquement sur MPI laisse peu à peu la place à des modèles mélangeant l'utilisation de threads et de MPI. Ce changement de modèle entraîne de nombreuses problématiques car les implémentations MPI n'ont pas été conçues pour supporter les applications multi-threadées. Dans cette thèse, afin de garantir le bon fonctionnement des communications, nous proposons un module logiciel faisant interagir l'ordonnanceur de threads et la bibliothèque de communication. Ce gestionnaire d'entrées/sorties générique prend en charge la détection des événements du réseau et exploite les multiples unités de calcul présentes sur la machine de manière transparente. Grâce à la collaboration étroite avec l'ordonnanceur de threads, le gestionnaire d'entrées/sorties que nous proposons assure un haut niveau de réactivité aux événements du réseau. Nous montrons qu'il est ainsi possible de faire progresser les communications réseau en arrière-plan et donc de recouvrir les communications par du calcul. La parallélisation de la bibliothèque de communication est également facilité par un mécanisme d'exportation de tâches capable d'exploiter les différentes unités de calcul disponible tout en prenant en compte la localité des données. Les gains obtenus sur des tests synthétiques et sur des applications montre que l'interaction entre la bibliothèque de communication et l'ordonnanceur de threads permet de réduire le coût des communications et donc d'améliorer les performances d'une application. / The current trend of constructors for scientific computation is to build clusters whose node include an increasing number of cores.The classical programming model that is only based on MPI is being replaced by hybrid approaches that mix communication and multi-threading. This evolution of the programming model leads to numerous problems since MPI implementations were not designed for multi-threaded applications. In this thesis, in order to guarantee a smooth behavior of communication, we propose a software module that interact with both the threads scheduler and the communication library. This module, by working closely with the thread scheduler, allows to make communication progress in the background and guarantees a high level of reactivity to network events, even when the node is overloaded. We show that this permits to make communication progress in the background and thus to overlap communication and computation. The parallelization of the communication library is also made easier thanks to a task onloading mechanism that is able to exploit the available cores while taking data locality into account. The results we obtain on synthetic application as well as real-life applications show that the interaction between the thread scheduler and the communication library allows to reduce the overhead of communication and thus to improve the application performance.

M. Tech (Department of Chemistry, Faculty of Applied and Computer Sciences) Vaal University of Technology. / A simple and rapid method for determination of estrone (E1) and β-estradiol (E2) was developed and validated using high performance liquid chromatography (HPLC). The solutions of standards and sample were prepared with distilled water. HPLC separation was performed in isocratic method 50/50 (water/methanol) using 4.6 mm x 250 mm id film thickness 5 µm) XDB-C18 capillary column, detector DAD, UV on 254 nm, temperature 20 ºC with flow rate of 2 mL/min, sample volume 20 µL and run time of 10 min. Calibration curves were linear between concentration range 1.0 - 15.0 ppm. The method was validated for limit of detection and quantification, linearity, precision, trueness and specificity. Also the method was applied to directly and easily to the analysis of the E1 and E2. Adsorption experiments were carried out in batch mode using multistirrer in a series of Erlenmeyer flasks of 50 ml capacity covered to prevent contamination having concentration ranges of E1 and E2 from 1 to 10 mg/L with adsorbent dose range 0.01 to 1 g at pH range 1 to 10 and temperature range 15°C to 35°C, placed on multistirrer. The results of the batch studies showed that simultaneous adsorption shows the maximum percent (91%) removal of E1 and (86 %) E2 at optimum temperature 25 °C of adsorbent dose 0.1 g, and pH 7. The mechanism, isotherms and kinetics of removal of two endocrine disrupting chemicals, estrone (E1) and β-estradiol (E2) by activated carbon adsorption were investigated in an agitated non-flow batch adsorption studies. Mathematical models were used to describe the adsorption phenomenon with the kinetic and thermodynamic parameters evaluated using the adsorption equilibrium data at varying temperatures. Higher adsorption rates were achieved at acidic to neutral pH ranges, with the sorption kinetic data showing a good fit to the pseudo second order rate equation and the Langmuir adsorption isotherm model for both E1 and E2. The Gibbs free energy were –16.68 kJ/mol and –17.34 kJ/mol for E1 and E2 respectively. The values of enthalpy for both E1 (84.50 kJ/mol) and E2 (90 kJ/mol) indicated a chemical nature of the sorption process. Both the isotherm and thermodynamic data obtained all supported the mechanism of adsorption of E1 and E2 to be mainly chemisorption’s supported by some physical attractions.

M. Tech (Department of Chemistry, Faculty of Applied and Computer Sciences) Vaal University of Technology. / In this study, analytical method for determining the chlorinated phenols in water was developed using High Performance Liquid Chromatography. The following four compounds which are 2, 4, 6- Trichlorophenol (2, 4, 6 TCP), 3-chlorophenol (3CP), 2, 4- Dichlorophenol (2, 4 DCP) and 4-chloro-3-methylphenol (4C3MP) were identified and quantified with a High Performance Liquid Chromatography (HPLC). The validation parameters tested were,: linearity, trueness, precision, detection limit of quantitation, sensitivity, specificity, selectivity. The linear calibration ranges of five standard solution from 1-10 ppm. The linearity ranges between 0.9298-0.9813. The activated carbon based on the waste macadamia nutshell activated carbon (MAC) was investigated for its potential uses as an adsorbent for chlorinated phenols removal and compared with grafted macadamia nutshell activated carbon (GMAC). The adsorbent was characterized with Fourier transform infrared spectrophotometer (FTIR), scanning electron microscope (SEM) and thermo gravimetric analysis (TGA). The parameters such as pH, temperature, contact time, concentration and adsorbent were investigated by adsorption technique. The strata C18E has been used before for the same reason and therefore the research was based on mimic the functional group of solid phase extraction (SPE) into macadamia activated carbon (MAC). The functional groups in SPE C18E are benzene and octadecyl. MAC was grafted with strata C18E functional groups to compare its potential with the SPE. The pseudo-first-order and pseudo-second-order kinetic models were applied to verify the experimental data. The pseudo-second order exhibited the best fit for the kinetic studies for MAC adsorption. Chemical removal of chlorinated phenols from wastewater is necessary to reduce harmful products on the environment and human health. Chlorinated phenols have been previously listed as some of the highest priority contaminants and as well as mainly important capability carcinogenic toxins released from chemical plants. Their availability in water supplies was perceived by their bad taste and smell. The acceptable chlorinated phenols concentration in portable water is 1 (mg/l) base on the approval of world health organization. The permanent checking of chlorinated phenols in environmental samples has a greater significance and stresses highly effectiveness, common selectively and great sensitively methods. The maximum uptake of Phenol using weighed mass of MAC was found to be 78 % and for GMAC was 84% for both 2,4,6TCP. t=250 min, pH=5, Co=1mg/l, T = 25 oC and m = 0.3 g/l were the optimum condition for Phenol-MAC system and GMAC system. Over all analysis of equilibrium model analysis indicates the fitness of Langmuir isotherm model to Phenol MAC adsorption system, suggesting a monolayer adsorption of phenol on the surface of MAC. Phenol adsorption capacity of MAC was found to be decreasing with increase in temperature suggesting that the adsorption process was exothermic in nature, which was further supported by the negative values of change in enthalpy. Characterization of MAC and GMAC confirmed the mesoporous texture, highly carbonaceous nature and a higher effective surface area of 912 m2/g. The highest phenol uptake capacity of GMAC was found to be 8.0049 mg/g. The optimal conditions for various process parameters are t = 250 min, pH=5, Co=1mg/l, T = 25 oC and m = 0.3 g/l were the optimum condition for Phenol-GMAC system. Like Phenol-MAC system, the kinetics studies confirmed that Phenol-GMAC adsorption system can be described by pseudo- second-order kinetics model. Equilibrium model analysis indicates the fitness of Langmuir isotherm model to Phenol-MAC adsorption system, suggesting a monolayer adsorption of phenol on the surface of GMAC. Phenol adsorption capacity of GMAC was found to be decreasing with increase in temperature suggesting that the adsorption process was exothermic in nature, which was further supported by the negative values of change in enthalpy. The negative values of Gibb’s free energy suggested that adsorption of phenol onto GMAC was a spontaneous process.

The work is given an analysis of Open Source Software (further referred as OSS), which allows use and create computer clusters. It explored the issue of clustering and construction of clusters. All installations, configuration and cluster management have been done on the operating system GNU / Linux. Presented OSS makes possible to compile a storage cluster, cluster with load distribution, cluster with high availability and computing cluster. Different types of benchmarks was theoretically analyzed, and practically used for measuring cluster’s performance. Results were compared with others, eg. the TOP500 list of the best clusters available online. Practical part of the work deals with comparing performance computing clusters. With several tens of computational nodes has been established cluster, where was installed package OpenMPI, which allows parallelization of calculations. Subsequently, tests were performed with the High Performance Linpack, which by calculation of linear equations provides total performance. Influence of the parallelization to algorithm PEA was also tested. To present practical usability, cluster has been tested by program John the Ripper, which serves to cracking users passwords. The work shall include the quantity of graphs clarifying the function and mainly showing the achieved results.

<p>In-situ workflows are a special class of scientific workflows, where different component applications (such as simulation, visualization, analysis) run concurrently, and data flows continuously between components during the whole workflow lifetime. Traditionally, simulations write large amounts of output data to persistent storage, which are later read for future analysis/visualization. In comparison, in-situ workflows allow analysis/visualization components to consume simulation data while the simulations are still running and thus reduce the I/O overhead. There are recent research works that focus on providing data transport libraries to help compose a group of applications into an integral in-situ workflow. However, only a few ``performance-oriented'' studies exist for in-situ workflows, and most of these works focus on workflows with simple structures (e.g., single producer and single consumer), also without consideration of heterogeneous environments for in-situ workflows. Being able to efficiently utilize heterogeneous computing resources such as multiple Clouds and HPCs can significantly accelerate real-world in-situ workflows, and benefit applications that require both significant computation power and real-time outputs(e.g., identifying abnormal patterns in fluid dynamics). The goal of this dissertation is to provide resource planning algorithms and runtime support, to improve in-situ workflow performance on heterogeneous environments.</p> <p><br></p> <p>This dissertation first investigates the emerging applications of in-situ workflows, which usually include parallel simulation, visualization, and analysis components. Two representative real-world in-situ workflows are studied in details-- a real-time CFD machine learning/visualization workflow and a wildfire spreading workflow. These workflows showcase the capability of in-situ workflows: e.g.,  decoupled and accelerated computation and fast near-real-time response time, however, there is a lack of resource planning and runtime support for general in-situ workflows. For resource planning, I first formulate the optimization problem, and then design and implement a heuristic algorithm called ``SNL'' (Scheduled-Neighbor-Lookup). SNL considers the pipelined execution pattern of in-situ workflows, and guides the resource planning of complex in-situ workflows to achieve higher workflow throughput. For the runtime support, I design and implement the ``INSTANT'' runtime framework, a runtime framework to configure, plan, launch, and monitor in-situ workflows for distributed computing environments. INSTANT provides intuitive interfaces to compose abstract in-situ workflows, manages in-site and cross-site data transfers with ADIOS2, and supports resource planning using profiled performance data. Experiments with the two use cases show that INSTANT can efficiently streamline the orchestration of complex in-situ workflows, and the resource planning capability allows INSTANT to plan and carry out fast workflow execution at different computing resource availabilities.</p>

No description available.

Purpose - ​The purpose of this thesis is to examine the organisational culture at the case company to find out whether or not Human lean can contribute to a high performance culture. Methodology - ​A qualitative study was conducted with a case study as the basis for investigating the company's organizational culture. The empirical data was collected by semi-structured interviews with the staff of the company. Findings - ​The empirical study first presents a summary based on the respondents' responses and thereafter displays the case company in comparison to the characteristic features and behaviours of high performance cultures. Lastly, the study shows howHuman lean, with the help of visual management, KASAM and continuous improvements, can contribute to a high performance culture. Contribution - ​This study tests existing theory in a new context and presents a scientific gap on how Human lean can contribute to a high performance culture. / Syfte - ​Syftet är att undersöka organisationskulturen på fallföretaget för att se om Human lean kan bidra till att skapa en high performance kultur. Metod - För att undersöka fallföretaget organisationskultur har en kvalitativ studie genomförts med fallstudie som grund. All empiri har utformats utifrån semistrukturerade intervjuer med personalen på fallföretaget. Resultat - I studiens empiri presenteras först en sammanställning utifrån respondenternas svar för att sedan jämföra hur fallföretaget förhåller sig till de karaktäristiska dragen och beteendena high performance culture erhåller. Avslutningsvis visarstudienhurHumanlean med hjälp av visuell styrning, KASAM och ständiga förbättringar bidrar till en high performance culture. Bidrag - Denna studie testar befintligt teori i ett nytt kontext och framför ett vetenskapligt gap om hur Human lean kan bidra till en high performance culture.

Increasing computational demand of simulations motivates the use of parallel computing systems. At the same time, this parallelism poses challenges to application developers. The Message Passing Interface (MPI) is a de-facto standard for distributed memory programming in high performance computing. However, its use also enables complex parallel programing errors such as races, communication errors, and deadlocks. Automatic tools can assist application developers in the detection and removal of such errors. This thesis considers tools that detect such errors during an application run and advances them towards a combination of both precise checks (neither false positives nor false negatives) and scalability. This includes novel hierarchical checks that provide scalability, as well as a formal basis for a distributed deadlock detection approach. At the same time, the development of parallel runtime tools is challenging and time consuming, especially if scalability and portability are key design goals. Current tool development projects often create similar tool components, while component reuse remains low. To provide a perspective towards more efficient tool development, which simplifies scalable implementations, component reuse, and tool integration, this thesis proposes an abstraction for a parallel tools infrastructure along with a prototype implementation. This abstraction overcomes the use of multiple interfaces for different types of tool functionality, which limit flexible component reuse. Thus, this thesis advances runtime error detection tools and uses their redesign and their increased scalability requirements to apply and evaluate a novel tool infrastructure abstraction. The new abstraction ultimately allows developers to focus on their tool functionality, rather than on developing or integrating common tool components. The use of such an abstraction in wide ranges of parallel runtime tool development projects could greatly increase component reuse. Thus, decreasing tool development time and cost. An application study with up to 16,384 application processes demonstrates the applicability of both the proposed runtime correctness concepts and of the proposed tools infrastructure.

Thesis (PhD)--Stellenbosch University, 2013. / ENGLISH ABSTRACT: Impact Polypropylene Copolymers (IPCs) are extremely complex materials, consisting of a mixture of polypropylene homopolymer and copolymers having different comonomer (ethylene) contents and chemical composition distributions. IPC can only be effectively analysed by multidimensional analytical approaches. For this, initially, the individual components have to be separated according to any of their molecular characteristics, either by chemical composition distribution (CCD) or molar mass distribution (MMD), followed by further analysis of these separated fractions with conventional analytical techniques. The combination of preparative temperature rising elution fractionation (TREF) with several other analytical techniques have been reported for the thorough characterization of this material. However, even the combinations of these methods were of limited value due to the complex nature of this polymer. Therefore, novel analytical approaches are needed for a more detailed compositional analysis of IPCs. This work describes a number of multidimensional analytical techniques that are based on the combination of fractionation and hyphenated techniques. Firstly, preparative TREF was combined with high temperature size exclusion chromatography-FTIR (HT SEC-FTIR), HT SEC-HPer DSC (High Performance Differential Scanning Calorimetry) and high temperature two-dimensional liquid chromatography (HT 2D-LC) for the comprehensive analysis of a typical impact polypropylene copolymer and one of its midelution temperature TREF fractions. HT SEC-FTIR analysis provided information regarding the chemical composition and crystallinity as a function of molar mass. Thermal analysis of selected SEC fractions using a novel DSC method - High Speed or High Performance Differential Scanning Calorimetry (HPer DSC) - that allows measuring of minute amounts of material down to micrograms, yielded the melting and crystallization behaviour of these fractions which is related to the chemical heterogeneity of this complex copolymer. High temperature 2D-LC analysis provided the complete separation of this TREF fraction according to the chemical composition of each component along with its molar mass distribution. In a second step, the compositional characterization by advanced thermal analysis (HPer DSC, Flash DSC 1, and solution DSC) of the TREF-SEC fractions was extended to all semi-crystalline and higher temperature TREF fractions. By applying HPer DSC at scan rates of 5−200°C/min and Flash DSC 1 at scan rates of 10−1000°C/s, the metastability of one of the fractions was studied in detail. DSC measurements of TREF-SEC cross-fractions at high scan rates in p-xylene successfully connected reversely to the slow scan rate in TREF elution, if corrected for recrystallization. Finally, the exact chemical structure of all HT HPLC separated components was determined by coupling of HT HPLC with FTIR spectroscopy via an LCTransform interface. This novel approach revealed the capability of this hyphenated technique to determine the exact chemical composition of the individual components in the complex TREF fractions of IPCs. The HT HPLC–FTIR results confirmed the separation mechanism in HPLC using a solvent gradient of 1-decanol/TCB and a graphitic stationary phase at 160°C. FTIR analysis provided information on the ethylene and propylene contents of the fractions as well as on the ethylene and propylene crystallinities. / AFRIKAANSE OPSOMMING: Impak Polipropileen Kopolimere (IPKe) is uiters komplekse materiale, bestaande uit 'n mengsel van polipropileen homopolimeer en kopolimere met verskillende komonomeer (etileen) inhoud en chemiese samestelling verspreiding. IPKe kan slegs doeltreffend ontleed word deur multidimensionele analitiese benaderings te volg. Hiervoor moet die individuele komponente aanvanklik eers geskei word volgens enige van hul molekulêre eienskappe, hetsy deur die chemiese samestelling verspreiding (CSV) of molêre massa verspreiding (MMV), gevolg deur 'n verdere ontleding van hierdie geskeide fraksies met konvensionele analitiese tegnieke. Die kombinasie van voorbereidings temperatuur-verhogings eluasie fraksionering (TVEF) met verskeie ander analitiese tegnieke is gerapporteer vir die deeglike karakterisering van hierdie materiaal. Maar selfs die kombinasies van hierdie metodes was van beperkte waarde as gevolg van die komplekse aard van hierdie polimeer. Daarom word nuwe analitiese benaderings benodig vir 'n meer gedetailleerde komposisionele ontleding van IPKe. Hierdie studie beskryf 'n aantal multidimensionele analitiese tegnieke wat gebaseer is op die kombinasie van fraksionering en gekoppelde tegnieke. Eerstens is voorbereidings TVEF gekombineer met hoë temperatuur grootte-uitsluitingschromatografie-FTIR (HT GUC-FTIR), HT GUC-HPer DSK en hoë temperatuur twee-dimensionele vloeistof chromatografie (HT 2D-VC) vir die omvattende ontleding van 'n tipiese impak polipropileen kopolimeer en een van sy mid-eluasie temperatuur TVEF fraksies. HT GUC-FTIR analiese het inligting verskaf met betrekking tot die chemiese samestelling en kristalliniteit as 'n funksie van molêre massa. Termiese analiese van geselekteerde GUC fraksies deur gebruik te maak van 'n nuwe-DSK metode - Hoë Spoed of Hoë Prestasie Differensïele skandeer kalorimetrie (HPer DSK) - wat die meting van klein hoeveelhede materiaal tot by mikrogram hoeveelhede toelaat, het die smelt en kristallisasie gedrag van hierdie fraksies bepaal wat verwant is aan die chemiese heterogeniteit van hierdie komplekse kopolimeer. Hoë temperatuur 2D-LC analiese het die volledige skeiding van hierdie TVEF fraksie volgens die chemiese samestelling van elke komponent saam met die molêre massa verspreiding moontlik gemaak. In 'n tweede stap, is die komposisionele karakterisering deur gevorderde termiese analiese (HPer DSK, Flash DSK 1 en oplossing DSK) van die TVEF-GUC fraksies uitgebrei na alle semi-kristallyne en hoër temperatuur TVEF fraksies. Deur die gebruik van HPer DSK, teen ’n skandeerspoed van 5-200°C / min, en Flash DSK 1, teen ’n skandeerspoed van 10-1000°C / s, is die meta-stabiliteit van een van die fraksies in detail bestudeer. DSK metings van TVEF-GUC kruis-fraksies by 'n hoë skandeeerspoed in p-xyleen het suksesvol omgekeerd verbind aan die stadige skandeerspoed in TVEF eluasie, wanneer gekorrigeer vir dekristallisatie. Ten slotte is die presiese chemiese struktuur van al die HT HPVC geskeide komponente bepaal deur die koppeling van HT HPVC met FTIR spektroskopie deur middel van 'n LC-transform-koppelvlak. Hierdie nuwe benadering het die vermoë van die gekoppelde tegniek om die presiese chemiese samestelling van die individuele komponente in die komplekse TVEF fraksies of IPKe te bepaal aan die lig gebring. Die HT HPVC-FTIR resultate het die skeidingsmeganisme in HPVC bevestig deur die gebruik van ’n oplosmiddelgradiënt van 1-dekanol/TCB en 'n graphitiese stasionêre fase by 160°C. FTIR analiese verskaf inligting in verband met die etileen en propileen inhoud van die fraksies sowel as die etileen en propileen krystalliniteit.

HPLC has been used and can quickly determine several ions simultaneously. The method of determination described for transition metals [Cr(III), Fe(III), Ni(II), Co(II), Cu(II), Zn(II), Cd(II), Mn(II)] and [Ca(II), Pb(II)] using HPLC with UV-VIS detection is better than the PAR complexation method commonly used. The effects of both eluent pH and detector wavelength were investigated. Results from using different pHs and wavelengths, optional analytical conditions for the separation of [Ni(II), Co(II), Cu(II)], [Cr(III), Fe(III), Ca(II), Ni(II), Cu(II)], and [Ca(II), Zn(II), Pb(II)] in one injection, respectively, are described. The influence of adding different concentrations of Na_2EDTA solvent to the sample is shown. Detection limits, linear range, and the comparisons between this study and a post-column PAR method are given.