No description available.

Thesis (M.S.)--University of Wisconsin--Madison, 1968. / eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references.

Thesis (Ph. D.)--University of Wisconsin--Madison, 1964. / Typescript. Vita. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references.

Due to the dependency of High-Level Architecture (HLA)-Based simulations on the resources of distributed environments, simulations can face load imbalances and can suffer from low performance in terms of execution time. HLA is a framework that simplifies the implementation of distributed simulations; it also has been built with dedicated resources in mind. As technology is nowadays shifting towards shared environments, the following two weaknesses have become apparent in HLA: managing federates and reacting towards load imbalances on shared resources. Moreover, a number of dynamic load management systems have been designed in order to provide a solution to enable a balanced simulation environment on shared resources. These solutions use specific techniques depending on simulation characteristics or load aspects to perform the balancing task. Load prediction is one of such techniques that improves load redistribution heuristics by preventing load imbalances. In this thesis, a number of enhancements for a prediction technique are presented, and their efficiency are compared. The proposed enhancements solve the observed problems with Holt’s implementations on dynamic load balancing systems for HLA-Based distributed simulations and provide better forecasting. As a result, these enhancements provide better predictions for the load oscillations of the shared resources. Furthermore, a number of federate migration decision-making approaches are introduced to add more intelligence into the process of migrating federates. The approaches aim to solve a dependency problem in the prediction-based load balancing system on the prediction model, thus making similar systems adapt to any future system improvements.

When player skill levels differ widely in a competitive First-Person Shooter (FPS) game, enjoyment suffers: weaker players become frustrated and stronger players become less engaged. Player balancing techniques attempt to assist the weaker player and make games more competitive, but these techniques have limitations for deployment when skill levels vary substantially. In this thesis, we developed new player balancing schemes to deal with a range of FPS skill difference, and tested these techniques in a series of five studies using a commercial-quality FPS game developed with the UDK engine. Our results showed that our balancing techniques (Combo and Delay) are extremely effective at balancing, even for players with large skill differences. These techniques also led to higher enjoyment of the game by players of all skill levels. Our studies are the first to show that player balancing can work well in realistic FPS games, providing developers with a way to increase the audience for this popular genre. In addition, our results demonstrate the idea that successful balancing is as much about the way the technique is applied as it is about the specific manipulation.

Game designers spend a great deal of time developing balanced game experiences. However, differences in player ability, hardware capacity (e.g. network connections) or real-world elements (as in mixed-reality games), make it difficult to balance games for different players in different conditions. In this research, adaptive time-variant minigames have been introduced as a method of addressing the challenges in time balancing as a part of balancing players of games. These minigames were parameterized to allow both a guaranteed minimum play time (the minimum time to complete a minigames to address the fixed temporal constraints) and dynamic adaptability (the ability of adapting the game during the game play to address temporal variations caused by individual differences). Three time adaptation algorithms have been introduced in this research and the interaction between adaptive algorithm, game mechanic, and game difficulty were analyzed in controlled experiments. The studies showed that there are significant effects and interactions for all three factors, confirming the initial hypothesis that these processes were important and linked to each other. Furthermore, the studies revealed that finer temporal granularity leads to less-perceptible adaptation and smaller deviations in game completion times. The results also provided evidence that adaptation mechanisms allow accurate prediction of play time. The designed minigames were valuable in helping to balance temporal asymmetries in a real mixed-reality game. It was also found that these adaptation algorithms did not interrupt the overall play experience.

Latent Semantic Indexing (LSI) is one of the popular techniques in the information retrieval fields. Different from the traditional information retrieval techniques, LSI is not based on the keyword matching simply. It uses statistics and algebraic computations. Based on Singular Value Decomposition (SVD), the higher dimensional matrix is converted to a lower dimensional approximate matrix, of which the noises could be filtered. And also the issues of synonymy and polysemy in the traditional techniques can be overcome based on the investigations of the terms related with the documents. However, it is notable that LSI suffers a scalability issue due to the computing complexity of SVD. This thesis presents a resource aware distributed LSI algorithm MR-LSI which can solve the scalability issue using Hadoop framework based on the distributed computing model MapReduce. It also solves the overhead issue caused by the involved clustering algorithm. The evaluations indicate that MR-LSI can gain significant enhancement compared to the other strategies on processing large scale of documents. One remarkable advantage of Hadoop is that it supports heterogeneous computing environments so that the issue of unbalanced load among nodes is highlighted. Therefore, a load balancing algorithm based on genetic algorithm for balancing load in static environment is proposed. The results show that it can improve the performance of a cluster according to heterogeneity levels. Considering dynamic Hadoop environments, a dynamic load balancing strategy with varying window size has been proposed. The algorithm works depending on data selecting decision and modeling Hadoop parameters and working mechanisms. Employing improved genetic algorithm for achieving optimized scheduler, the algorithm enhances the performance of a cluster with certain heterogeneity levels.

Ho Kin Chuen Matthew. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1998. / Includes bibliographical references (leaves 153-157). / Abstract also in Chinese. / Chapter 1 --- Introduction --- p.15 / Chapter 1.1 --- The U-line Balancing Problem --- p.15 / Chapter 1.2 --- Configuration of an U-line --- p.17 / Chapter 1.3 --- Feasible subsets and sequences --- p.19 / Chapter 1.4 --- Assignment of tasks to stations --- p.21 / Chapter 1.5 --- Costs --- p.22 / Chapter 1.6 --- Formulation of The U-line Balancing Problem --- p.23 / Chapter 1.7 --- Design of computational study --- p.25 / Chapter 1.7.1 --- Input parameters --- p.25 / Chapter 1.7.2 --- Output variables --- p.26 / Chapter 1.7.3 --- Problems solved --- p.27 / Chapter 1.7.3.1 --- Problem Set One --- p.28 / Chapter 1.7.3.2 --- Problem Set Two --- p.28 / Chapter 1.7.3.3 --- Problem Set Three --- p.29 / Chapter 1.7.3.4 --- Problem Set Four --- p.29 / Chapter 1.8 --- Contributions --- p.29 / Chapter 1.9 --- Organization of thesis --- p.30 / Chapter 2 --- Literature Review --- p.31 / Chapter 2.1 --- Introduction --- p.31 / Chapter 2.2 --- The Straight-line Balancing Problem --- p.32 / Chapter 2.2.1 --- Single-model Assembly Line Balancing with deterministic task time (SMD) --- p.34 / Chapter 2.2.2 --- Single-model Assembly Line Balancing with stochastic task times (SMS) --- p.36 / Chapter 2.2.3 --- Multi/Mixed-model Assemble Line Balancing with deterministic task times (MMD) --- p.37 / Chapter 2.2.4 --- Multi/Mixed-model Assembly Line Balancing with stochastic task times (MMS) --- p.38 / Chapter 2.3 --- The U-line Balancing Problem --- p.39 / Chapter 2.4 --- Conclusions --- p.45 / Chapter 3 --- Heuristic Methods --- p.47 / Chapter 3.1 --- Introduction --- p.47 / Chapter 3.2 --- Single-pass heuristic methods --- p.47 / Chapter 3.3 --- Computational results --- p.50 / Chapter 3.3.1 --- Problem Set One --- p.50 / Chapter 3.3.2 --- Problem Set Two --- p.52 / Chapter 3.3.3 --- Problem Set Three --- p.54 / Chapter 3.3.4 --- Problem Set Four --- p.55 / Chapter 3.4 --- Discussions --- p.57 / Chapter 3.5 --- Conclusions --- p.59 / Chapter 4 --- Genetic Algorithm --- p.60 / Chapter 4.1 --- Introduction --- p.60 / Chapter 4.2 --- Application of GA to The Straight-line Balancing Problem --- p.61 / Chapter 4.3 --- Application of GA to The U-line Balancing Problem --- p.62 / Chapter 4.3.1 --- Coding scheme --- p.63 / Chapter 4.3.2 --- Initial population --- p.64 / Chapter 4.3.3 --- Fitness function --- p.65 / Chapter 4.3.4 --- Selection scheme --- p.66 / Chapter 4.3.5 --- Reproduction --- p.67 / Chapter 4.3.6 --- Replacement scheme --- p.68 / Chapter 4.3.7 --- Elitism --- p.68 / Chapter 4.3.8 --- Termination criteria --- p.68 / Chapter 4.4 --- Repair method --- p.69 / Chapter 4.5 --- Crossover operators --- p.71 / Chapter 4.5.1 --- Sequence and configuration infeasible crossover operators --- p.72 / Chapter 4.5.1.1 --- Partially Mapped Crossover (PMX) --- p.72 / Chapter 4.5.1.2 --- Order Crossover #1 (ORD#l) --- p.74 / Chapter 4.5.1.3 --- Order Crossover #2 (ORD#2) --- p.74 / Chapter 4.5.1.4 --- Position Based Crossover (POS) --- p.75 / Chapter 4.5.1.5 --- Cycle Crossover (CYC) --- p.76 / Chapter 4.5.1.6 --- Edge Recombination Crossover (EDG) --- p.77 / Chapter 4.5.1.7 --- Enhanced Edge Recombination Crossover (EEDG) --- p.80 / Chapter 4.5.1.8 --- Uniform-order Based Crossover (UOX) --- p.81 / Chapter 4.5.2 --- Sequence feasible but configuration infeasible crossover operators --- p.82 / Chapter 4.5.2.1 --- One-point Crossover (1PX) --- p.82 / Chapter 4.5.2.2 --- Two-point Crossover (2PX) --- p.84 / Chapter 4.5.2.3 --- Uniform Crossover (UX) --- p.85 / Chapter 4.6 --- Mutation operators --- p.86 / Chapter 4.6.1 --- Sequence infeasible mutation operators --- p.87 / Chapter 4.6.1.1 --- Inversion (INV) --- p.87 / Chapter 4.6.1.2 --- Insertion (INS) --- p.87 / Chapter 4.6.1.3 --- Displacement (DIS) --- p.88 / Chapter 4.6.1.4 --- Reciprocal Exchange (RE) --- p.88 / Chapter 4.6.2 --- Sequence and configuration feasible mutation operators --- p.89 / Chapter 4.6.2.1 --- Scramble Mutation (SCR) --- p.89 / Chapter 4.6.2.2 --- Feasible Insertion (FINS) --- p.90 / Chapter 4.7 --- Computational study --- p.91 / Chapter 4.7.1 --- Comparison of crossover operators --- p.91 / Chapter 4.7.2 --- Comparison of mutation operators --- p.95 / Chapter 4.7.2.1 --- Order crossover#2 and mutation operators --- p.95 / Chapter 4.7.2.2 --- Position based crossover and mutation operators --- p.97 / Chapter 4.7.3 --- Parameters setting --- p.99 / Chapter 4.7.4 --- Computational results --- p.104 / Chapter 4.7.5 --- Comparative results --- p.105 / Chapter 4.7.5.1 --- Problem Set One --- p.105 / Chapter 4.7.5.2 --- Problem Set Two --- p.105 / Chapter 4.7.5.3 --- Problem Set Three --- p.107 / Chapter 4.7.5.4 --- Problem Set Four --- p.107 / Chapter 4.8 --- Conclusions --- p.109 / Chapter 5 --- Dynamic Programming and Lower Bounds --- p.110 / Chapter 5.1 --- Dynamic Programming (DP) --- p.110 / Chapter 5.1.1 --- Introduction --- p.110 / Chapter 5.1.2 --- Modified Dynamic Programming algorithm --- p.112 / Chapter 5.1.3 --- Comparison between optimal solution and heuristics --- p.120 / Chapter 5.1.4 --- Comparison between optimal solution and the GA --- p.123 / Chapter 5.2 --- Lower Bounds --- p.123 / Chapter 5.2.1 --- Introduction --- p.123 / Chapter 5.2.2 --- The U-line Balancing Problem and The Bin Packing Problem --- p.127 / Chapter 5.2.3 --- Martello and Toth's lower bounds for The BPP --- p.128 / Chapter 5.2.3.1 --- Bound L1 --- p.128 / Chapter 5.2.3.2 --- Bound L2 --- p.128 / Chapter 5.2.3.3 --- Dominances and reductions --- p.129 / Chapter 5.2.3.3.1 --- Dominance criterion --- p.129 / Chapter 5.2.3.3.2 --- Reduction procedure --- p.130 / Chapter 5.2.3.4 --- Lower Bound LR --- p.131 / Chapter 5.2.4 --- Chen and Srivastava's lower bounds for The BPP --- p.131 / Chapter 5.2.4.1 --- A unified lower bound --- p.132 / Chapter 5.2.4.2 --- Improving Lm --- p.133 / Chapter 5.2.4.3 --- "Computing a lower bound on N(1/4,1]" --- p.134 / Chapter 5.2.5 --- Lower bounds for The U-line Balancing Problem --- p.137 / Chapter 5.2.5.1 --- Lower bounds on number of stations required --- p.137 / Chapter 5.2.5.2 --- Lower bounds on total cost --- p.139 / Chapter 5.2.6 --- Computational results --- p.140 / Chapter 5.2.6.1 --- Results for different Problem Sets --- p.140 / Chapter 5.2.6.2 --- Comparison between lower bounds and optimal solutions --- p.143 / Chapter 5.2.6.3 --- Comparison between lower bounds and heuristics --- p.145 / Chapter 5.2.6.4 --- Comparison between lower bounds and GA --- p.147 / Chapter 5.3 --- Conclusions --- p.149 / Chapter 6 --- Conclusions --- p.150 / Chapter 6.1 --- Summary of achievements --- p.150 / Chapter 6.2 --- Future works --- p.151

Balancing an air distribution system consists primarily of measuring airflow and adjusting volume control devices to get specified airflow. Flow calculation methods are not accurate enough to ensure proper balancing by duct design alone. To assure proper balancing, dampers within an air distribution system must be adjusted until design flows are met throughout the system to within + 10%. By properly balancing an air distribution system, operating costs in the system will be reduced, comfort for the occupants in the building will be increased, and the life of the HVAC equipment will be improved. Existing balancing techniques are iterative methods that require several measurements and damper adjustments. The flows and pressures are first measured, and then the dampers are adjusted to match design airflow. The flows are measured again and the measuring and adjusting process is repeated on a trial-and-error process until design flow is achieved. This iterative process is time consuming and expensive. The proposed new balancing technique is to use a computer program that, based on a few measurements, determines damper adjustments that will achieve design airflow throughout the system. Each terminal damper is adjusted only once to a specified flowrate that is determined by the computer program, making the balancing process quicker and less expensive. No iteration is required. / Master of Science

The performance of communication and storage systems can be improved if the data being sent or stored has certain patterns and structure. In particular, some benefit if the frequency of the symbols is balanced. This includes magnetic and optical data storage devices, as well as future holographic storage systems. Significant research has been done to develop techniques and algorithms to adapt the data (in a reversible manner) to these systems. The goal has been to restructure the data to improve performance while keeping the complexity as low as possible. In this thesis, we consider balancing binary sequences and present its application in holographic storage systems. An overview is given of different approaches, as well as a survey of previous balancing methods. We show that common compression algorithms can be used for this purpose both alone and combined with other balancing algorithms. Simplified models are analyzed using information theory to determine the extent of the compression in this context. Simulation results using standard data are presented as well as theoretical analysis for the performance of the combination of compression with other balancing algorithms. / Graduate