No description available.

<p> Hybrid task-lists are sets of periodic and asynchronous processes. To verifiably schedule hybrid tasks-lists with hard and soft real-time requirements, Xu and Lam proposed Integrated Pre-Run-Time scheduling (IPRTS) [13], a two phase method that first makes use of pre-run-time scheduling techniques, converting some asynchronous tasks with hard deadlines to periodic tasks and reserving processor capacity for the remaining hard deadline asynchronous tasks. These remaining asynchronous tasks are scheduled by a novel run-time scheduler that enforces arbitrary exclusion relations between any combination of periodic and asynchronous processes. The technique has two significant drawbacks: (i) a custom run-time scheduler is required that is not available on existing Real-Time Operating Systems (RTOS) and (ii) in many circumstances the reservation of processor capacity is overly pessimistic, causing the failure of the method for many simple task lists. To overcome these drawbacks, this thesis narrows the set of task-lists considered to those where the asynchronous tasks exclude periodic tasks and periodic processes do not exclude asynchronous tasks. A high priority polling server is then used to handle all hard asynchronous tasks. In cases where the method succeeds, it is easily implementable on any RTOS that has priority based scheduling with phased release times, and inherits the error handling and soft real-time process scheduling capabilities of the RTOS. A set of software tools which partially automates the technique, including an open source implementation of the Xu-Parnas pre-run-time scheduling algorithm [14], has been developed and applied to the examples in the thesis.</p> / Thesis / Master of Applied Science (MASc)

<p> This thesis defends the use of genetic algorithms (GA) to solve the maximum number of repetitions in a binary string. Repetitions in strings have significant uses in many different fields, whether it is data-mining, pattern-matching, data compression or computational biology 14]. Main extended the definition of repetition, he realized that in some cases output could be reduced because of overlapping repetitions, that are simply rotations of one another [10]. As a result, he designed the notion of a run to capture the maximal leftmost repetition that is extended to the right as much as possible. Franek and Smyth independently computed the same number of maximum repetition for strings of length five to 35 using an exhaustive search method. Values greater than 35 were not computed because of the exponential increase in time required. Using GAs we are able to generate string with very large, if not the maximum, number of runs for any string length. The ability to generate strings with large runs is an advantage for learning more about the characteristics of these strings. </p> / Thesis / Master of Science (MSc)

Reconfigurable computing architectures are gaining popularity as a replacement for general-purpose architectures for many high performance embedded applications. These machines support parallel computation and direct the data from the producers of an intermediate result to the consumers over custom pathways. The Wormhole Run-time Reconfigurable (RTR) computing architecture is a concept developed at Virginia Tech to address the weaknesses of contemporary FPGAs for configurable computing. The Stallion chip is a full-custom configurable computing "FPGA"-like integrated circuit with a coarse grained nature. Based on the result of the first generation device, the Colt chip, the Stallion chip is a follow-up configurable computing chip. This thesis focuses on the VLSI layout implementation of the Stallion chip. Effort has been made to explain many facts and advantages of the Wormhole Configurable Computing Machine (CCM). Design techniques, strategies, circuit characterization, performance estimation, and ways to solve problems when using CAD layout design tools are illustrated. / Master of Science

Thesis (MEng)--Stellenbosch University, 2015. / ENGLISH ABSTRACT: Over the last few decades, wave run up prediction has gained the interest of numerous researchers and every newly-published paper has aimed to predict wave run up with greater accuracy. Wave run up is defined as the vertical elevation reached by a wave's, front water edge as it runs up a beach, measured relative to the still water line. Wave run up is dependent on the incidental wave height, the wave period, the beach slope and the wave steepness. The majority of publications incorporate all of these factors, but some do not, which has led to numerous debates. The goal of this study is to do a re-assessment of previously published wave run up formulae, to obtain a more informed understanding about wave run up and the available predictive empirical formulae. The study also seeks to evaluate the Mather, Stretch & Garland (2011) formula. The method for undertaking this objective comprised a physical model test series with 10 regular wave conditions on a constant slope, being 1/24, performed with an impermeable floor. Also, a beach study in the field was done on Long Beach, Noordhoek, where run up measurements were taken for 30 minute intervals, resulting in five test conditions. A numerical model was employed in conjunction with the beach study to determine the local offshore wave parameters transformed from a deep water wave rider. This information was used to correlate the run up measurements with known wave parameters. Firstly, the physical model assessment was performed to provide a proper foundation for run up understanding. Plotting empirical normalised run up values (R2/H0 ) versus the Iribarren number for different formulae, a grouping was achieved with upper and lower boundaries. The physical model results plotted on the lower end of this grouping, resulted in prediction differences of more than 10%. These differences may have been caused by the unevenness of the physical model slope or the fact that only one slope had been tested. Despite this, the results fell within a band of wave run up formulae located on the lower end of this grouping. An assessment of the beach measurements in the field gave a better correlation than the physical model results when compared to normalised predicted wave run up formulae. These measurements also plotted on the lower end of the grouping, resulting in prediction differences of less than 10% for some empirical formulae. When comparing these empirical predictions to one another, the results demonstrate that the formulae comparing best with the beach measurements were Holman (1986) and Stockdon, Holman, Howd, & Sallenger Jr. (2006). Extreme over predictions were found by Mase & Iwagaki (1984), Hedges & Mase (2004) and Douglass (1992). Nielsen & Hanslow (1991) only compared best with the beach measurements and De la Pena, Sanchez Gonzalez, Diaz-Sanchez, & Martin Huescar (2012) only compared best to the physical model results. This study supports the formula proposed by Mather, Stretch, & Garland (2011). Applying their formula to the measured results presented a C constant of 3.3 for the physical model and 8.6 for the beach results. Both values are within the range prescribed by the authors. Further reasearch minimized the array of possible „C‟ values by correlating this coefficient to Iribarren numbers. „C‟ values between 3.0~5.0 is prescribed for low Iribarren conditions (0.25-0.4) and values between 7.0~10 for higher Iribarren conditions are 0.75-0.8. However, this formula is still open for operator erros whereby the „C‟ value has a big influence in the final result. The best formulae to use, from results within this thesis, is proposed by Holman (1986) and Stockdon et.al (2006). These formulae are not open to operator erros and uses the significant wave height, deep water wave length and the beach face slope to calculate the wave run up. / AFRIKAANSE OPSOMMING: Gedurende die afgelope paar dekades, het golf-oploop voorspellings die aandag van talle navorsers gelok en elke nuwe geskrewe voorlegging het gepoog om meer akkurate golf-oploop voorspellings te verwesenlik. golf-oploop kan definieer word as die vertikale elevasie bereik deur 'n golf se voorwaterkant soos dit op die strand uitrol, gemeet relatief vanaf die stilwaterlyn. golf-oploop is afhanklik van die invals-golfhoogte, die golfperiode, die strandhelling en die golfsteilheid. Die oorgrote mederheid publikasies uit die literaturr inkorporeer al hierdie faktore, maar sommige nie, wat groot debatvoering tot gevolg het. Die doel met hierdie studie is om vorige gepubliseerde golf- oploop formules te re-evalueer, om 'n meer ingeligte begrip van golf- oploop en beskikbare voorspellende formules te verkry. Die studie poog terselfdertyd ook om golf-opvolg tendense, uniek aan Suid Afrikaanse strande te evalueer deur die huidige formule wat tans hier gebruik word, te assesseer. Om hierdie doelwit te bereik, is gebruik gemaak van 'n fisiese model toets reeks bestaande uit 10 reëlmatige golfstoestande op 'n konstante ondeurlaatbaare strandhelling van 1/24. 'n Veldstudie was ook uitgevoer op Langstrand, Noordhoek, waar golf-oploopmetings met 30 minute tussenposes uitgevoer is, vir vyf toets-toestande. Tesame met die veldstudie, is 'n numeriese model aangewend om die gemete diepsee data nader ann die strand wat bestudeer is te transformeer. Hierdie inligting is benodig om 'n verband tussen tussen oploop-metings en bekende golf parameters te bepaal. Eerstens is die fisiese model assessering uitgevoer om 'n behoorlike basis vir die begrip van golfoploop in die veld te verkry. Deur die emperiese, genormaliseerde oploop waardes (R₂/H₀) vir verkeie formules teenoor die Iribarren getal te plot, is 'n groepering met hoër en laer grense gevind. Daar is gevind dat die fisiese modelwaardes op die laer grens plot, en het verskille met die emperiese waardes van meer as 10% getoon. Hierdie verskille is moontlik veroorsaak as gevolg van 'n oneweredige fisiese model strandhelling of deur die feit dat slegs een helling getoets is. Ten spyte hiervan, het die model oploop waardes binne die bestek van golf- oploop formules geval. Assessering van die veldmetings het 'n beter korrelasie as die fisiese modelresultate getoon, tydens vergelykings met genormaliseerde golf-oploop formules van die emperiese formules. Die oploop waardes van hierdie metings het ook geplot aan die laer grens van die groepering, met verskille van minder as 10% vir die meeste gevalle van die emperiese formules. Wanneer hierdie emperiese voorspellings vergelyk word, is gevind dat die formules wat die beste ooreenstem met die fisiese model, die van Holman (1986) en Stockdon, Howd, & Sallenger Jr. (2006) is. Die emperiese formules van Mase & Iwagake (1984), Hedges & Mase (2004) en Douglas (1992) het die golf-oploop oorvoorspel. Nielsen & Hanslow (1991) het slegs die beste met die strandmetings vergelyk, terwyl De la Pena, Sanchez Gonzalez, Diaz-Sanchez & Martin Huescar (2012) slegs die beste vergelyk het met die fisiese-model resultaat. Hierdie studie ondersteun die formule voorgestel deur Mather, Stretch, & Garland (2011). Deur hul formules op die gemete bevindings toe te pas, is 'n C konstante van 3.3 vir die fisiese model resultate, en 8.0 vir die stranduitlslae bepaal. Beide waardes lê binne die grense wat deur die outeurs voorgestel is. Verdere navorsing het getoon dat moontlike waardes vir die „C‟ konstante tussen 3.0 en 5.0 moet wees vir Iribarren waardes van tussen 0.25 en 0.4. Vir hoër Iribarren waardes, 0.75-0.8, moet die „C‟ kosntante tussen 7.0 en 10 wees; dog is die formule steeds oop vir operateur foute. Die hoofbevindinge van die tesis is gevind dat die beste golf-oploop formules, om tans te gebruik, die van Holman (1986) en Stockdon et.al (2006) is. Hierdie formules kan glad nie beinvloed word deur operateurs foute nie en maak gebruik van die invals golfhoogte, die golfperiode en die strandhelling om die golf-oploop te bepaal.

An investigation of the genetic and meristic patterns of coastal cutthroat trout (Oncorhynchus clarki clarki) in two coastal basins, one in Oregon and one in Alaska, revealed varying degrees of differences among populations within each site. Coastal cutthroat trout collected from seven sites above and below barriers from Elk River, Oregon were characterized by significant genetic structuring (Fst=0.095) based on allozymes and significant differences among nine meristic characters. This may reflect the geologic history and local conditions of the Elk River Basin. Despite this divergence, Elk River populations were relatively similar to each other (Nei's genetic identity> 0.0987), suggesting that these populations share common ancestry. Fewer differences were detected among the coastal cutthroat trout from four sites below and above a barrier within Vixen Inlet, Alaska. Genetic structuring among populations was low (Fst=0.016) and significant differences were detected in only two of nine meristic characters. This may reflect the recent glacial history of southeast Alaska. In both sites, there was concordance between allozyme data and meristic data. / Graduation date: 1996

Title: Basic anthropometric characteristics of 400 m and 800 m runers. Student: Bc. Vendulka Praská Supervisor: PaedDr. Jitka Vindušková, CSc. Objectives: The main objective of this dissertation is comparison of anthropometric characteristics of 400m and 800m runners. Another objective is analysis of relation between somatometric parameters and physical performance of 400m and 800m runners. Methods: In this thesis was used comparison method of athletes' body characteristics using graphs and correlations of anthropometric characteristics with physical performance. Results: We found, that average body height and weight of 400m run athletes is 1,84 m and 73 kg for men and 1,70 m and 60 kg for women. Average body height and weight of 800m run athletes is 1,77 m and 65 kg for men and 1,66 m and 53 kg for women. Furthermore we found by using statistical calculations that there is a direct relation between body height, weight and physical performance. Keywords: athletics, anthropometric characteristics, 400 m run, 800 m run

Quantifying spatial and temporal patterns of rapid channelized erosion, on human time scales, is critical to understanding its processes and their consequences. This investigation utilized field observations, repeat terrestrial laser scanning (TLS), and Structure-from-Motion photogrammetry (SfM) to document the size and retreat rates of a knickpoint, defined as a localized near-vertical reach of a fluvial channel, and its contribution to erosion, in an urbanizing landscape with a loess substrate. The Bull Mountain area, in Washington County, southwest of Portland, Oregon, is an ideal study area, offering a measurable knickpoint that translates the response of the rapid erosion throughout this transient system. Previous urbanization there has increased peak flows in streams, potentially initiating rapid channel incision and associated slope instability and sediment pollution, affecting real property and infrastructure. Despite the documented increase in discharge, upstream migration rates of the knickpoint, as well as the overall channel erosion rate, were unknown. Sequential point cloud analysis quantified topographic changes in the landscape, in three dimensions, throughout time. The measured minimum knickpoint migration rates ranged from - 0.23 m/yr to - 2.45 m/yr with an average of - 1.52 m/yr and minimum of total volume eroded of 6.49 m3. The negative sign indicates the upstream direction. An extreme erosion event caused - 12.5 m of erosion in ~ 4.5 months. The interval including the extreme erosion event was recorded separately using traditional measurement techniques and resulted in an average retreat rate of - 4.31 m/yr. Analysis of patterns of erosion revealed four primary modes: exfoliation, large soil block failure, undercutting at the knickpoint base, and upper bank failure. Results from soil analyses indicate a layer of high bulk density (1.85 g/cm3) loess at the base of the upper channel may restrict the channelized incision for that reach and control the height and geometry of the knickpoint face, leading to a parallel mode of retreat. From the observed erosion rates a substrate specific average value of erodibility, or K value, of 0.01 m0.2 yr-1, was determined. As erosion forces the retreat of these knickpoints upstream, the effects of increasing urban runoff are felt throughout the watershed. The work presented here provides insight on the physical controls driving erosion and can serve as a prologue for future mitigation.

Bull Run Watershed was set aside in late the 1800s as the water supply source for the City of Portland. Other than two dams being constructed, Reservoir #1 (1929) and Reservoir #2 (1962), development of the land had been minimal as public access was restricted. In the early 1960s, land management changed with increased road building and timber removal raising concerns about increased sediment discharge into the reservoirs. The objective of this study is to evaluate how much and how fast the sediment has accumulated in Reservoir #1, and to determine if the rate of sediment accumulation has changed over time. Three methods are utilized: 1) differencing map comparing pre- and postimpoundment sediment conditions, 2) analysis of tree-stumps on reservoir floor, and 3) gravity coring of reservoir sediment. Combining these methods, sediment volume is estimated between 254,000-422,000 cubic meters (332,000-552,000 cubic yards) and the rate of accumulation between 11.5-19.1 tonnes/km2/yr, reflecting a relatively low sediment yield rate. Two anomalous event-layers were identified in gravity cores collected. These are interpreted to be the 1964 flood and the 1972 North Fork Slide. Using these two events, sediment yield rate was divided into different historical segments: 15.33 (1930-1965); 43.62 (1965-1972); and 17.00 tonnes/km2/yr (1972-1993). The increase from 1965-1972 is attributed to either residual affects from the 1964 flood and/or changes in land management activities during this time. The source of the reservoir sediment is primarily from upper tributaries, with 20 percent being attributed to the anomalous events. Smaller amounts of sediment come from the reservoir side walls as lake levels raise and lower. Suspension and turbidity conditions in the reservoir are affected by the dynamics of the drainage system including seasonal fluctuations. Turbidity remains high at the upper reaches of the reservoir before settling out closer to the dam. Some sediment possibly leaves the reservoir over the spill-way or when water is removed for power production.

This study investigates the linkage between long-run and short-run dynamics of exchange rate determination for the German mark/U.S. dollar quarterly rate for the period 1973-1990. Earlier investigations failed to explicitly take into account the possible nonstationarity of the data set they were using. This study continues the work performed in this area by applying modern econometric techniques to empirical tests of the Dornbusch model. In essence, this study revives the monetary model and determines if the empirical analysis using the German/U.S. case derives elements which are compatible with the monetary theory of exchange rate determination.