Optimising the management of a heterogeneous stand

Abstract

Data-parallel computation in parallel and distributed environments

The past few decades have seen a considerable increase in the number of parallel and distributed systems. With the development of more complex applications, the need for more powerful systems has emerged and various parallel and distributed environments have been designed and implemented. Each of the environments, including hardware and software, has unique strengths and weaknesses. There is no single parallel environment that can be identified as the best environment for all applications with respect to hardware and software properties. The main goal of this thesis is to provide a novel way of performing data-parallel computation in parallel and distributed environments by utilizing the best characteristics of difference aspects of parallel computing. For the purpose of this thesis, three aspects of parallel computing were identified and studied. First, three parallel environments (shared memory, distributed memory, and a network of workstations) are evaluated to quantify theirsuitability for different parallel applications. Due to the parallel and distributed nature of the environments, networks connecting the processors in these environments were investigated with respect to their performance characteristics. Second, scheduling algorithms are studied in order to make them more efficient and effective. A concept of application-specific information scheduling is introduced. The application- specific information is data about the workload extractedfrom an application, which is provided to a scheduling algorithm. Three scheduling algorithms are enhanced to utilize the application-specific information to further refine their scheduling properties. A more accurate description of the workload is especially important in cases where the workunits are heterogeneous and the parallel environment is heterogeneous and/or non-dedicated. The results obtained show that the additional information regarding the workload has a positive impact on the performance of applications. Third, a programming paradigm for networks of symmetric multiprocessor (SMP) workstations is introduced. The MPIT programming paradigm incorporates the Message Passing Interface (MPI) with threads to provide a methodology to write parallel applications that efficiently utilize the available resources and minimize the overhead. The MPIT allows for communication and computation to overlap by deploying a dedicated thread for communication. Furthermore, the programming paradigm implements an application-specific scheduling algorithm. The scheduling algorithm is executed by the communication thread. Thus, the scheduling does not affect the execution of the parallel application. Performance results achieved from the MPIT show that considerable improvements over conventional MPI applications are achieved.

MCMC Analysis Of Classical Time Series Algorithms

Identification of order of an Autoregressive Moving Average Model (ARMA) by the usual graphical method is subjective. Hence, there is a need of developing a technique to identify the order without employing the graphical investigation of series autocorrelations. To avoid subjectivity, this thesis focuses on determining the order of the Autoregressive Moving Average Model using Reversible Jump Markov Chain Monte Carlo (RJMCMC). The RJMCMC selects the model from a set of the models suggested by better fitting, standard deviation errors and the frequency of accepted data. Together with deep analysis of the classical Box-Jenkins modeling methodology the integration with MCMC algorithms has been focused through parameter estimation and model fitting of ARMA models. This helps to verify how well the MCMC algorithms can treat the ARMA models, by comparing the results with graphical method. It has been seen that the MCMC produced better results than the classical time series approach.

On Energy Efficient Computing Platforms

In accordance with the Moore's law, the increasing number of on-chip integrated transistors has enabled modern computing platforms with not only higher processing power but also more affordable prices. As a result, these platforms, including portable devices, work stations and data centres, are becoming an inevitable part of the human society. However, with the demand for portability and raising cost of power, energy efficiency has emerged to be a major concern for modern computing platforms. As the complexity of on-chip systems increases, Network-on-Chip (NoC) has been proved as an efficient communication architecture which can further improve system performances and scalability while reducing the design cost. Therefore, in this thesis, we study and propose energy optimization approaches based on NoC architecture, with special focuses on the following aspects. As the architectural trend of future computing platforms, 3D systems have many bene ts including higher integration density, smaller footprint, heterogeneous integration, etc. Moreover, 3D technology can signi cantly improve the network communication and effectively avoid long wirings, and therefore, provide higher system performance and energy efficiency. With the dynamic nature of on-chip communication in large scale NoC based systems, run-time system optimization is of crucial importance in order to achieve higher system reliability and essentially energy efficiency. In this thesis, we propose an agent based system design approach where agents are on-chip components which monitor and control system parameters such as supply voltage, operating frequency, etc. With this approach, we have analysed the implementation alternatives for dynamic voltage and frequency scaling and power gating techniques at different granularity, which reduce both dynamic and leakage energy consumption. Topologies, being one of the key factors for NoCs, are also explored for energy saving purpose. A Honeycomb NoC architecture is proposed in this thesis with turn-model based deadlock-free routing algorithms. Our analysis and simulation based evaluation show that Honeycomb NoCs outperform their Mesh based counterparts in terms of network cost, system performance as well as energy efficiency.

Scalable algorithms for height field illumination

Global illumination algorithms are at the center of realistic image synthesis and account for non-trivial light transport and occlusion within scenes, such as indirect illumination, ambient occlusion, and environment lighting. Their computationally most difficult part is determining light source visibility at each visible scene point. Height fields, on the other hand, constitute an important special case of geometry and are mainly used to describe certain types of objects such as terrains and to map detailed geometry onto object surfaces. The geometry of an entire scene can also be approximated by treating the distance values of its camera projection as a screen-space height field. In order to shadow height fields from environment lights a horizon map is usually used to occlude incident light. We reduce the per-receiver time complexity of generating the horizon map on N N height fields from O(N) of the previous work to O(1) by using an algorithm that incrementally traverses the height field and reuses the information already gathered along the path of traversal. We also propose an accurate method to integrate the incident light within the limits given by the horizon map. Indirect illumination in height fields requires information about which other points are visible to each height field point. We present an algorithm to determine this intervisibility in a time complexity that matches the space complexity of the produced visibility information, which is in contrast to previous methods which scale in the height field size. As a result the amount of computation is reduced by two orders of magnitude in common use cases. Screen-space ambient obscurance methods approximate ambient obscurance from the depth bu er geometry and have been widely adopted by contemporary real-time applications. They work by sampling the screen-space geometry around each receiver point but have been previously limited to near- field effects because sampling a large radius quickly exceeds the render time budget. We present an algorithm that reduces the quadratic per-pixel complexity of previous methods to a linear complexity by line sweeping over the depth bu er and maintaining an internal representation of the processed geometry from which occluders can be efficiently queried. Another algorithm is presented to determine ambient obscurance from the entire depth bu er at each screen pixel. The algorithm scans the depth bu er in a quick pre-pass and locates important features in it, which are then used to evaluate the ambient obscurance integral accurately. We also propose an evaluation of the integral such that results within a few percent of the ray traced screen-space reference are obtained at real-time render times.

Influence of reactant absorption and dissolution in heterogeneous precipitation processes

In this research work, the aim was to investigate the volumetric mass transfer coefficient [kLa] of oxygen in stirred tank in the presence of solid particle experimentally. The kLa correlations as a function of propeller rotation speed and flow rate of gas feed were studied. The O2 and CO2 absorption in water and in solid-liquid suspensions and heterogeneous precipitation of MgCO3 were thoroughly examined. The absorption experiments of oxygen were conducted in various systems like pure water and in aqueous suspensions of quartz and calcium carbonate particles. Secondly, the precipitation kinetics of magnesium carbonate was also investigated. The experiments were performed to study the reactive crystallization with magnesium hydroxide slurry and carbon dioxide gas by varying the feed rates of carbon dioxide and rotation speeds of mixer. The results of absorption and precipitation are evaluated by titration, total carbon (TC analysis), and ionic chromatrography (IC). For calcium carbonate, the particle concentration was varied from 17.4 g to 2382 g with two size fractions: 5 µm and 45-63 µm sieves. The kLa and P/V values of 17.4 g CaCO3 with particle size of 5µm and 45-63 µm were 0.016 s-1 and 2400 W/m3. At 69.9 g concentration of CaCO3, the achieved kLa is 0.014 s-1 with particle size of 5 µm and 0.017 s-1 with particle size of 45 to 63 µm. Further increase in concentration of calcium carbonate, i.e. 870g and 2382g , does not affect volumetric mass transfer coeffienct of oxygen. It could be concluded from absorption results that maximum value of kLa is 0.016 s-1. Also particle size and concentration does affect the transfer rate to some extend. For precipitation experiments, the constant concentration of Mg(OH)2 was 100 g and the rotation speed varied from 560 to 750 rpm, whereas the used feed rates of CO2 were 1 and 9 L/min. At 560 rpm and feed rate of CO2 is 1 L/min, the maximum value of Mg ion and TC were 0.25 mol/litre and 0.12 mol/litre with the residence time of 40 min. When flow rate of CO2 increased to 9 L/min with same 560 rpm, the achieved value of Mg and TC were 0.3 mol/litre and 0.12 mol/L with shorter residence time of 30 min. It is concluded that feed rate of CO2 is dominant in precipitation experiments and it has a key role in dissociation and reaction of magnesium hydroxide in precipitation of magnesium carbonate.

Efficient Optimization Algorithms for Nonlinear Data Analysis

Identification of low-dimensional structures and main sources of variation from multivariate data are fundamental tasks in data analysis. Many methods aimed at these tasks involve solution of an optimization problem. Thus, the objective of this thesis is to develop computationally efficient and theoretically justified methods for solving such problems. Most of the thesis is based on a statistical model, where ridges of the density estimated from the data are considered as relevant features. Finding ridges, that are generalized maxima, necessitates development of advanced optimization methods. An efficient and convergent trust region Newton method for projecting a point onto a ridge of the underlying density is developed for this purpose. The method is utilized in a differential equation-based approach for tracing ridges and computing projection coordinates along them. The density estimation is done nonparametrically by using Gaussian kernels. This allows application of ridge-based methods with only mild assumptions on the underlying structure of the data. The statistical model and the ridge finding methods are adapted to two different applications. The first one is extraction of curvilinear structures from noisy data mixed with background clutter. The second one is a novel nonlinear generalization of principal component analysis (PCA) and its extension to time series data. The methods have a wide range of potential applications, where most of the earlier approaches are inadequate. Examples include identification of faults from seismic data and identification of filaments from cosmological data. Applicability of the nonlinear PCA to climate analysis and reconstruction of periodic patterns from noisy time series data are also demonstrated. Other contributions of the thesis include development of an efficient semidefinite optimization method for embedding graphs into the Euclidean space. The method produces structure-preserving embeddings that maximize interpoint distances. It is primarily developed for dimensionality reduction, but has also potential applications in graph theory and various areas of physics, chemistry and engineering. Asymptotic behaviour of ridges and maxima of Gaussian kernel densities is also investigated when the kernel bandwidth approaches infinity. The results are applied to the nonlinear PCA and to finding significant maxima of such densities, which is a typical problem in visual object tracking.

Heterogeneous mass transfer in fluidized beds by computational fluid dynamics

The main objective of this research is to estimate and characterize heterogeneous mass transfer coefficients in bench- and pilot-scale fluidized bed processes by the means of computational fluid dynamics (CFD). A further objective is to benchmark the heterogeneous mass transfer coefficients predicted by fine-grid Eulerian CFD simulations against empirical data presented in the scientific literature. First, a fine-grid two-dimensional Eulerian CFD model with a solid and gas phase has been designed. The model is applied for transient two-dimensional simulations of char combustion in small-scale bubbling and turbulent fluidized beds. The same approach is used to simulate a novel fluidized bed energy conversion process developed for the carbon capture, chemical looping combustion operated with a gaseous fuel. In order to analyze the results of the CFD simulations, two one-dimensional fluidized bed models have been formulated. The single-phase and bubble-emulsion models were applied to derive the average gas-bed and interphase mass transfer coefficients, respectively. In the analysis, the effects of various fluidized bed operation parameters, such as fluidization, velocity, particle and bubble diameter, reactor size, and chemical kinetics, on the heterogeneous mass transfer coefficients in the lower fluidized bed are evaluated extensively. The analysis shows that the fine-grid Eulerian CFD model can predict the heterogeneous mass transfer coefficients quantitatively with acceptable accuracy. Qualitatively, the CFD-based research of fluidized bed process revealed several new scientific results, such as parametrical relationships. The huge variance of seven orders of magnitude within the bed Sherwood numbers presented in the literature could be explained by the change of controlling mechanisms in the overall heterogeneous mass transfer process with the varied process conditions. The research opens new process-specific insights into the reactive fluidized bed processes, such as a strong mass transfer control over heterogeneous reaction rate, a dominance of interphase mass transfer in the fine-particle fluidized beds and a strong chemical kinetic dependence of the average gas-bed mass transfer. The obtained mass transfer coefficients can be applied in fluidized bed models used for various engineering design, reactor scale-up and process research tasks, and they consequently provide an enhanced prediction accuracy of the performance of fluidized bed processes.

α-Pinene oxide and verbenol oxide isomerizations over heterogeneous catalysts

Terpenes are a valuable natural resource for the production of fine chemicals. Turpentine, obtained from biomass and also as a side product of softwood industry, is rich in monoterpenes such as α-pinene and β-pinene, which are widely used as raw materials in the synthesis of flavors, fragrances and pharmaceutical compounds. The rearrangement of their epoxides has been thoroughly studied in recent years, as a method to obtain compounds which are further used in the fine chemical industry. The industrially most desired products of α-pinene oxide isomerization are campholenic aldehyde and trans-carveol. Campholenic aldehyde is an intermediate for the manufacture of sandalwood-like fragrances such as santalol. Trans-carveol is an expensive constituent of the Valencia orange essence oil used in perfume bases and food flavor composition. Furthermore it has been found to exhibit chemoprevention of mammary carcinogenesis. A wide range of iron and ceria supported catalysts were prepared, characterized and tested for α-pinene oxide isomerization in order to selective synthesis of above mentioned products. The highest catalytic activity in the preparation of campholenic aldehyde over iron modified catalysts using toluene as a solvent at 70 °C (total conversion of α-pinene oxide with a selectivity of 66 % to the desired aldehyde) was achieved in the presence of Fe-MCM-41. Furthermore, Fe-MCM-41 catalyst was successfully regenerated without deterioration of catalytic activity and selectivity. The most active catalysts in the synthesis of trans-carveol from α-pinene oxide over iron and ceria modified catalysts in N,N-dimethylacetamide as a solvent at 140 °C (total conversion of α-pinene oxide with selectivity 43 % to trans-carveol) were Fe-Beta-300 and Ce-Si-MCM-41. These catalysts were further tested for an analogous reaction, namely verbenol oxide isomerization. Verbenone is another natural organic compound which can be found in a variety of plants or synthesized by allylic oxidation of α-pinene. An interesting product which is synthesized from verbenone is (1R,2R,6S)-3-methyl-6-(prop-1-en-2-yl)cyclohex-3-ene-1,2-diol. It has been discovered that this diol possesses potent anti-Parkinson activity. The most effective way leading to desired diol starts from verbenone and includes three stages: epoxidation of verbenone to verbenone oxide, reduction of verbenone oxide and subsequent isomerization of obtained verbenol oxide, which is analogous to isomerization of α-pinene oxide. In the research focused on the last step of these synthesis, high selectivity (82 %) to desired diol was achieved in the isomerization of verbenol oxide at a conversion level of 96 % in N,N-dimethylacetamide at 140 °C using iron modified zeolite, Fe-Beta-300. This reaction displayed surprisingly high selectivity, which has not been achieved yet. The possibility of the reuse of heterogeneous catalysts without activity loss was demonstrated.

Hardware Accelarated Visual Tracking Algorithms. A Systematic Literature Review

Many industrial applications need object recognition and tracking capabilities. The algorithms developed for those purposes are computationally expensive. Yet ,real time performance, high accuracy and small power consumption are essential measures of the system. When all these requirements are combined, hardware acceleration of these algorithms becomes a feasible solution. The purpose of this study is to analyze the current state of these hardware acceleration solutions, which algorithms have been implemented in hardware and what modifications have been done in order to adapt these algorithms to hardware.

3D Mesh Simplification. A survey of algorithms and CAD model simplification tests

Simplification of highly detailed CAD models is an important step when CAD models are visualized or by other means utilized in augmented reality applications. Without simplification, CAD models may cause severe processing and storage is- sues especially in mobile devices. In addition, simplified models may have other advantages like better visual clarity or improved reliability when used for visual pose tracking. The geometry of CAD models is invariably presented in form of a 3D mesh. In this paper, we survey mesh simplification algorithms in general and focus especially to algorithms that can be used to simplify CAD models. We test some commonly known algorithms with real world CAD data and characterize some new CAD related simplification algorithms that have not been surveyed in previous mesh simplification reviews.

Efficient Algorithms for Coastal Geographic Problems

The increasing performance of computers has made it possible to solve algorithmically problems for which manual and possibly inaccurate methods have been previously used. Nevertheless, one must still pay attention to the performance of an algorithm if huge datasets are used or if the problem iscomputationally difficult. Two geographic problems are studied in the articles included in this thesis. In the first problem the goal is to determine distances from points, called study points, to shorelines in predefined directions. Together with other in-formation, mainly related to wind, these distances can be used to estimate wave exposure at different areas. In the second problem the input consists of a set of sites where water quality observations have been made and of the results of the measurements at the different sites. The goal is to select a subset of the observational sites in such a manner that water quality is still measured in a sufficient accuracy when monitoring at the other sites is stopped to reduce economic cost. Most of the thesis concentrates on the first problem, known as the fetch length problem. The main challenge is that the two-dimensional map is represented as a set of polygons with millions of vertices in total and the distances may also be computed for millions of study points in several directions. Efficient algorithms are developed for the problem, one of them approximate and the others exact except for rounding errors. The solutions also differ in that three of them are targeted for serial operation or for a small number of CPU cores whereas one, together with its further developments, is suitable also for parallel machines such as GPUs.