Energy efficiency is of the utmost importance in modern high-performance embedded processor design. As the number of transistors on a chip continues to increase each year, and processor logic becomes ever more complex, the dynamic switching power cost of running such processors increases. The continual progression in fabrication processes brings a reduction in the feature size of the transistor structures on chips with each new technology generation. This reduction in size increases the significance of leakage power (a constant drain that is proportional to the number of transistors). Particularly in embedded devices, the proportion of an electronic product’s power budget accounted for by the CPU is significant (often as much as 50%). Dynamic branch prediction is a hardware mechanism used to forecast the direction, and target address, of branch instructions. This is essential to high performance pipelined and superscalar processors, where the direction and target of branches is not computed until several stages into the pipeline. Accurate branch prediction also acts to increase energy efficiency by reducing the amount of time spent executing mis-speculated instructions. ‘Stalling’ is no longer a sensible option when the significance of static power dissipation is considered. Dynamic branch prediction logic typically accounts for over 10% of a processor’s global power dissipation, making it an obvious target for energy optimisation. Previous approaches at increasing the energy efficiency of dynamic branch prediction logic has focused on either fully dynamic or fully static techniques. Dynamic techniques include the introduction of a new cache-like structure that can decide whether branch prediction logic should be accessed for a given branch, and static techniques tend to focus on scheduling around branch instructions so that a prediction is not needed (or the branch is removed completely). This dissertation explores a method of combining static techniques and profiling information with simple hardware support in order to reduce the number of accesses made to a branch predictor. The local delay region is used on unconditional absolute branches to avoid prediction, and, for most other branches, Adaptive Branch Bias Measurement (through profiling) is used to assign a static prediction that is as accurate as a dynamic prediction for that branch. This information is represented as two hint-bits in branch instructions, and then interpreted by simple hardware logic that bypasses both the lookup and update phases for appropriate branches. The global processor power saving that can be achieved by this Combined Algorithm is around 6% on the experimental architectures shown. These architectures are based upon real contemporary embedded architecture specifications. The introduction of the Combined Algorithm also significantly reduces the execution time of programs on Multiple Instruction Issue processors. This is attributed to the increase achieved in global prediction accuracy.

As early as 1994, and with transformation in mind, the African National Congress (ANC) identified the development of the previously disadvantaged communities in South Africa as a national priority. In so doing, sport, among others, was used in this strategy as a medium to improve and change the circumstances of people in disadvantaged communities. Sports development in disadvantaged communities is essential, if taken into consideration that without such further support of the sport talent of some of these children, the needs of these talented children cannot be met and their potential will remain undeveloped. Talent identification (TID), which is based on scientific principles and forms the first step in sports development, still is relatively new in South Africa. Historically, coaches used their own knowledge and experience of the characteristics, which should lead to success in sport as well as participation in competitions in order to do TID. To date, very little research has been done on talent identification and development in sprints and long-jump, especially pertaining to young boy and girl athletes, and in particular concerning athletes from disadvantaged communities. The first and second objectives of this study were to implement sport specific athletics development programmes aimed at improving sprinting and long-jump ability and to determine its effect on the abilities and skills of talented 10 to 15 year-old girls and boys with talent for sprints and long-jump. The third and fourth objectives of this study were to establish which kinanthropometric, physical and motor components will play such a role in 10 to 15 year-olds that it can predict performance ability in sprints and long-jump in girls and boys at this age. The "Australian Talent Search" protocol was used to identify general sport talent in the children (66 girls and 62 boys) who were identified for the study. The talented children (19 girls and 21 boys) from the initial group of were then subjected to a sport specific test battery for sprints and long-jump. The maturation level of the boys was determined by means of a maturity questionnaire, based on the 5 Tanner stages. By using the Statistica and SAS computer programmes, independent t-testing, covariance analyses, correlation coefficients, effect sizes, descriptive statistics as well as a stepwise multiple regression analysis were used to analyze the data with regard to the above-mentioned objectives. A p-value smaller than or equal to 0.05 was accepted as significant. From the results of the study it is evident that the development programmes contributed to the improvement of physical and motor abilities and skills important for performance in both sprints and long-jump in girls (n=19) and boys (n=21) respectively. Among the girls, flexibility, explosive power, muscle endurance, reaction time, speed, speed endurance, acceleration and long-jump showed statistically significant improvement, while abdominal muscle strength and stride length showed no improvement. Secondly, it was established that the development programme contributed statistically significantly to an improvement in flexibility, muscle endurance, 0-40 metres speed and long-jump ability in boys. However, some components did not show improvement, among them explosive power, reaction time, speed endurance, acceleration and stride length. A stepwise multiple regression analysis was used to analyze the third and fourth objectives. It indicated that, in respect to the 100 metres sprint, 7 variables, namely long-jump, push-ups to the point of exhaustion, 7-level abdominal strength, 0-5 metres speed, ankle dorsiflexion, body length and age contributed 84.0% to the total variance in girls. As for long-jump, 7 variables, namely 0-100 metres speed, body length, 7-level abdominal strength, push-ups, ankle dorsiflexion, standing long-jump and body mass proved to be the most important contributors to performance in these items with a total contribution of 79% to the total variance. The fourth objective indicated that average anaerobic power output, acceleration and body mass contributed statistically significantly to performance in the 100 metres sprint in boys with a contribution of 86.5% to the total variance. Horizontal jump, age and acceleration contributed statistically significantly to long-jump performance with a contribution of 81 5% to the total variance. It is evident from this study that sport specific development programmes can successfully be implemented on girls and boys at ages 10-15 in order to improve sprinting and long-jump ability, regardless of poverty-stricken circumstances and poor infrastructure. Furthermore, the study brought to light that specific kinanthropometric, physical and motor abilities exist which can be used to predict performance in sprints and long-jump in girls and boys separately at ages 10 to 15. / Thesis (Ph.D. (Human Movement Science))--North-West University, Potchefstroom Campus, 2007

Title: Refine the prediction process for the software Combatfit Aim: Find more accurate method to predict for the need of software Combatfit. Compare software prediction with field measurements and suitable alternative for predicting. The methods used: The thesis is empiric - teoretic research and it is consist of two parts. 1) Teoretic part - analytic work 2) Empiric part - metodologic study Results: The nomogram was chosen as a suitable way of prediction. The research was realized at 3 km, 8 km and 10 km on track in the field. The measured times were used to compare the predicted time by software Combatfit with time prediction according to the nomogram. It was found that the nomogram predictions outperforms software Combatfit. Furthermore it was determined tightness of the relationship between predicted and measured time in either process. The tightness of the relationship with the nomogram was very high up to absolute. The tightness of the relationship with the software Comatfit was moderate. This statement applies to test files that was used. It is not entirely clear whether this would achieve the same conclusions for other performance groups. Key words: Software Combatfit, prediction of performance, movement, external conditions, nomograme.

Thesis (Ph.D. (Human Movement Science))--North-West University, Potchefstroom Campus, 2007.

As early as 1994, and with transformation in mind, the African National Congress (ANC) identified the development of the previously disadvantaged communities in South Africa as a national priority. In so doing, sport, among others, was used in this strategy as a medium to improve and change the circumstances of people in disadvantaged communities. Sports development in disadvantaged communities is essential, if taken into consideration that without such further support of the sport talent of some of these children, the needs of these talented children cannot be met and their potential will remain undeveloped. Talent identification (TID), which is based on scientific principles and forms the first step in sports development, still is relatively new in South Africa. Historically, coaches used their own knowledge and experience of the characteristics, which should lead to success in sport as well as participation in competitions in order to do TID. To date, very little research has been done on talent identification and development in sprints and long-jump, especially pertaining to young boy and girl athletes, and in particular concerning athletes from disadvantaged communities. The first and second objectives of this study were to implement sport specific athletics development programmes aimed at improving sprinting and long-jump ability and to determine its effect on the abilities and skills of talented 10 to 15 year-old girls and boys with talent for sprints and long-jump. The third and fourth objectives of this study were to establish which kinanthropometric, physical and motor components will play such a role in 10 to 15 year-olds that it can predict performance ability in sprints and long-jump in girls and boys at this age. The "Australian Talent Search" protocol was used to identify general sport talent in the children (66 girls and 62 boys) who were identified for the study. The talented children (19 girls and 21 boys) from the initial group of were then subjected to a sport specific test battery for sprints and long-jump. The maturation level of the boys was determined by means of a maturity questionnaire, based on the 5 Tanner stages. By using the Statistica and SAS computer programmes, independent t-testing, covariance analyses, correlation coefficients, effect sizes, descriptive statistics as well as a stepwise multiple regression analysis were used to analyze the data with regard to the above-mentioned objectives. A p-value smaller than or equal to 0.05 was accepted as significant. From the results of the study it is evident that the development programmes contributed to the improvement of physical and motor abilities and skills important for performance in both sprints and long-jump in girls (n=19) and boys (n=21) respectively. Among the girls, flexibility, explosive power, muscle endurance, reaction time, speed, speed endurance, acceleration and long-jump showed statistically significant improvement, while abdominal muscle strength and stride length showed no improvement. Secondly, it was established that the development programme contributed statistically significantly to an improvement in flexibility, muscle endurance, 0-40 metres speed and long-jump ability in boys. However, some components did not show improvement, among them explosive power, reaction time, speed endurance, acceleration and stride length. A stepwise multiple regression analysis was used to analyze the third and fourth objectives. It indicated that, in respect to the 100 metres sprint, 7 variables, namely long-jump, push-ups to the point of exhaustion, 7-level abdominal strength, 0-5 metres speed, ankle dorsiflexion, body length and age contributed 84.0% to the total variance in girls. As for long-jump, 7 variables, namely 0-100 metres speed, body length, 7-level abdominal strength, push-ups, ankle dorsiflexion, standing long-jump and body mass proved to be the most important contributors to performance in these items with a total contribution of 79% to the total variance. The fourth objective indicated that average anaerobic power output, acceleration and body mass contributed statistically significantly to performance in the 100 metres sprint in boys with a contribution of 86.5% to the total variance. Horizontal jump, age and acceleration contributed statistically significantly to long-jump performance with a contribution of 81 5% to the total variance. It is evident from this study that sport specific development programmes can successfully be implemented on girls and boys at ages 10-15 in order to improve sprinting and long-jump ability, regardless of poverty-stricken circumstances and poor infrastructure. Furthermore, the study brought to light that specific kinanthropometric, physical and motor abilities exist which can be used to predict performance in sprints and long-jump in girls and boys separately at ages 10 to 15. / Thesis (Ph.D. (Human Movement Science))--North-West University, Potchefstroom Campus, 2007

Following the growth of high performance computing systems (HPC) in size and complexity, and the advent of faster and more complex Exascale systems, failures became the norm rather than the exception. Hence, the protection mechanisms need to be improved. The most de facto mechanisms such as checkpoint/restart or redundancy may also fail to support the continuous operation of future HPC systems in the presence of failures. Failure prediction is a new protection approach that is beneficial for HPC systems with a short mean time between failure. The failure prediction mechanism extends the existing protection mechanisms via the dynamic adjustment of the protection level. This work provides a prototype to analyze and predict system behavior using statistical analysis to pave the path toward resilience in HPC systems. The proposed anomaly detection method is noise-tolerant by design and produces accurate results with as little as 30 minutes of historical data. Machine learning models complement the main approach and further improve the accuracy of failure predictions up to 85%. The fully automatic unsupervised behavior analysis approach, proposed in this work, is a novel solution to protect future extreme-scale systems against failures.:1 Introduction 1.1 Background and Statement of the Problem 1.2 Purpose and Significance of the Study 1.3 Jam–e Jam: A System Behavior Analyzer 2 Review of the Literature 2.1 Syslog Analysis 2.2 Users and Systems Privacy 2.3 Failure Detection and Prediction 2.3.1 Failure Correlation 2.3.2 Anomaly Detection 2.3.3 Prediction Methods 2.3.4 Prediction Accuracy and Lead Time 3 Data Collection and Preparation 3.1 Taurus HPC Cluster 3.2 Monitoring Data 3.2.1 Data Collection 3.2.2 Taurus System Log Dataset 3.3 Data Preparation 3.3.1 Users and Systems Privacy 3.3.2 Storage and Size Reduction 3.3.3 Automation and Improvements 3.3.4 Data Discretization and Noise Mitigation 3.3.5 Cleansed Taurus System Log Dataset 3.4 Marking Potential Failures 4 Failure Prediction 4.1 Null Hypothesis 4.2 Failure Correlation 4.2.1 Node Vicinities 4.2.2 Impact of Vicinities 4.3 Anomaly Detection 4.3.1 Statistical Analysis (frequency) 4.3.2 Pattern Detection (order) 4.3.3 Machine Learning 4.4 Adaptive resilience 5 Results 5.1 Taurus System Logs 5.2 System-wide Failure Patterns 5.3 Failure Correlations 5.4 Taurus Failures Statistics 5.5 Jam-e Jam Prototype 5.6 Summary and Discussion 6 Conclusion and Future Works Bibliography List of Figures List of Tables Appendix A Neural Network Models Appendix B External Tools Appendix C Structure of Failure Metadata Databse Appendix D Reproducibility Appendix E Publicly Available HPC Monitoring Datasets Appendix F Glossary Appendix G Acronyms

The web and mobile web traffic is growing rapidly due to service accessibility in densely populated areas and massive usage of smart hand held devices (smartphones, pads and Internet dongles). This is a challenge for mobile operators and service providers, because it forces systematically increasing network throughput capacity and in parallel to keep this investment on profitable basis. On the other side, the user satisfaction with mobile broadband service should be considered as well, in order to balance the user expectations and mobile operator investments. Our work provides extensive QoE study for different conditions in sense of mobile web service, web content, network conditions and end user device. Finally, the obtained results create basis for more dimensional overview of mobile web QoE and allow to recognize quality and saturation thresholds based on network parameters. Further, obtained data provide base for design three prediction models for prediction of MOS (one for notebook, one for mobile device) and acceptability.