Automatic physical database design tools rely on ``what-if'' interfaces to the query optimizer to estimate the execution time of the training query workload under different candidate physical designs. The tools use these what-if interfaces to recommend physical designs that minimize the estimated execution time of the input training workload. Minimizing estimated execution time alone can lead to designs that are not robust to query optimizer errors and workload changes. In particular, if the optimizer makes errors in estimating the execution time of the workload queries, then the recommended physical design may actually degrade the performance of these queries. In this sense, the physical design is risky. Furthermore, if the production queries are slightly different from the training queries, the recommended physical design may not benefit them at all. In this sense, the physical design is not general. We define Risk and Generality as two new measures aimed at evaluating the robustness of a proposed physical database design, and we show how to extend the objective function being optimized by a generic physical design tool to take these measures into account. We have implemented a physical design advisor in PostqreSQL, and we use it to experimentally demonstrate the usefulness of our approach. We show that our two new metrics result in physical designs that are more robust, which means that the user can implement them with a higher degree of confidence. This is particularly important as we move towards truly zero-administration database systems in which there is not the possibility for a DBA to vet the recommendations of the physical design tool before applying them.

Automatic physical database design tools rely on ``what-if'' interfaces to the query optimizer to estimate the execution time of the training query workload under different candidate physical designs. The tools use these what-if interfaces to recommend physical designs that minimize the estimated execution time of the input training workload. Minimizing estimated execution time alone can lead to designs that are not robust to query optimizer errors and workload changes. In particular, if the optimizer makes errors in estimating the execution time of the workload queries, then the recommended physical design may actually degrade the performance of these queries. In this sense, the physical design is risky. Furthermore, if the production queries are slightly different from the training queries, the recommended physical design may not benefit them at all. In this sense, the physical design is not general. We define Risk and Generality as two new measures aimed at evaluating the robustness of a proposed physical database design, and we show how to extend the objective function being optimized by a generic physical design tool to take these measures into account. We have implemented a physical design advisor in PostqreSQL, and we use it to experimentally demonstrate the usefulness of our approach. We show that our two new metrics result in physical designs that are more robust, which means that the user can implement them with a higher degree of confidence. This is particularly important as we move towards truly zero-administration database systems in which there is not the possibility for a DBA to vet the recommendations of the physical design tool before applying them.

Human-like performance by machines in tasks of speech and audio processing has remained an elusive goal. In an attempt to bridge the gap in performance between humans and machines there has been an increased effort to study and model physiological processes. However, the widespread use of biologically inspired features proposed in the past has been hampered mainly by either the lack of robustness across a range of signal-to-noise ratios or the formidable computational costs. In physiological systems, sensor processing occurs in several stages. It is likely the case that signal features and biological processing techniques evolved together and are complementary or well matched. It is precisely for this reason that modeling the feature extraction processes should go hand in hand with modeling of the processes that use these features. This research presents a front-end feature extraction method for audio signals inspired by the human peripheral auditory system. New developments in the field of machine learning are leveraged to build classifiers to maximize the performance gains afforded by these features. The structure of the classification system is similar to what might be expected in physiological processing. Further, the feature extraction and classification algorithms can be efficiently implemented using the low-power cooperative analog-digital signal processing platform. The usefulness of the features is demonstrated for tasks of audio classification, speech versus non-speech discrimination, and speech recognition. The low-power nature of the classification system makes it ideal for use in applications such as hearing aids, hand-held devices, and surveillance through acoustic scene monitoring

A methodology for the conceptual design phase risk assessment of an aerospace system was proposed. The method was designed to examine political, social, and economic risk over a systems lifecycle through the use of future scenarios to bound uncertainty. A decision support framework was developed to allow the user to visualize the differences in performance and economic metrics between design options as well as allowing the user to visualize the effects of mitigating certain risks. A historical proof of concept was developed to test the methodology. The results indicated that the new method will work to examine political, social, and economic risk during conceptual level design, and that this information can be used to aid in design down-selection and decision making. The use of scenario-based analysis as an alternative to traditional probabilistic analysis allowed for better traceability and bounding of uncertainty. Other findings regarding the use of a risk analysis early during concept design and future work are also discussed.

When using data for estimating the parameters in a bivariate distribution, the tradition is to assume that data comes from a bivariate normal distribution. If the distribution is not bivariate normal, which often is the case, the maximum likelihood (ML) estimators are intractable and the least square (LS) estimators are inefficient. Here, we consider two independent sets of bivariate data which come from non-normal populations. We consider two distinctive distributions: the marginal and the conditional distributions are both Generalized Logistic, and the marginal and conditional distributions both belong to the Student&rsquo / s t family. We use the method of modified maximum likelihood (MML) to find estimators of various parameters in each distribution. We perform a simulation study to show that our estimators are more efficient and robust than the LS estimators even for small sample sizes. We develop hypothesis testing procedures using the LS and the MML estimators. We show that the latter are more powerful and robust. Moreover, we give a comparison of our tests with another well known robust test due to Tiku and Singh (1982) and show that our test is more powerful. The latter is based on censored normal samples and is quite prominent (Lehmann, 1986). We also use our MML estimators to find a more efficient estimator of Mahalanobis distance. We give real life examples.

In quality control applications, Time-Between-Events (TBE) type observations may be monitored by using Exponentially Weighted Moving Average (EWMA) control charts. A widely accepted model for the TBE processes is the exponential distribution, and hence TBE EWMA charts are designed under this assumption. Nevertheless, practical applications do not always conform to the theory and it is common that the observations do not fit the exponential model. Therefore, control charts that are robust to departures from the assumed distribution are desirable in practice. In this thesis, robustness of the lower-sided TBE EWMA charts to the assumption of exponentially distributed observations has been investigated. Weibull and lognormal distributions are considered in order to represent the departures from the assumed exponential model and Markov Chain approach is utilized for evaluating the performance of the chart. By analyzing the performance results, design settings are suggested in order to achieve robust lower-sided TBE EWMA charts.

For many years, incompatible computer-aided design (CAD) packages that are based on Non-uniform Rational B-Spline (NURBS) technology carried out the exchange of models and data through either neutral file formats (IGES or STEP) or proprietary formats that have been accepted as quasi industry standards. Although it is the only available solution at the current time, the exchange process most often produces unsatisfactory results. Models that are impeccable in the original modeling system usually end up with gaps or intersections between surfaces on another incompatible system. Issues such as loss of information, change of data accuracy, inconsistent tolerance, and misinterpretation of the original design intent are a few examples of problems associated with migrating models between different CAD systems. While these issues and drawbacks are well known and cost the industry billions of dollars every year, a solution to eradicate problems from their sources has not been developed. Meanwhile, researchers along with the industries concerned with these issues have been trying to resolve such problems by finding means to repair the migrated models either manually or by using specialized software. Designing in recent years is becoming more knowledge intensive and it is essential for NURBS to take its share of the ever increasing use of knowledge. NURBS are very powerful modeling tools and have become the de facto standard in modeling. If we stretch their strength and make them knowledge driven, benefits beyond current expectations can be achieved easily. This dissertation introduces knowledge guided NURBS with theoretical and practical foundations for supporting design intent capturing, retrieval, and exchange among dissimilar CAD systems. It shows that if NURBS entities are tagged with some knowledge, we can achieve seamless data exchange, increase robustness, and have more reliable computations, all of which are ultimate objectives many researchers in the field of CAD have been trying to accomplish for decades. Establishing relationships between a NURBS entity and its origin and destinations can aid with seamless CAD model migration. The type of the NURBS entity and the awareness of any irregularities can lead to more intelligent decisions on how to proceed with many computations to increase robustness and achieve a high level of reliability. As a result, instead of having models that are hardly modifiable because of migrating raw numerical data in isolation, the knowledge driven migration process will produce models that are editable and preserve design intent. We have addressed the issues not only theoretically but also by developing a prototype system that can serve as a test bed. The developed system shows that a click of a button can regenerate a migrated model instead of repairing it, avoiding delay and corrective processes that only limit the effective use of such models.

Urban drainage systems are influenced by several future drivers that affect the performance as well as the costs of the systems. The uncertainties associated with future drivers and their impact creates difficulties in designing urban drainage systems sustainably. A review of the different future drivers for urban drainage systems illustrates that no sufficient future predictions for the long operational life spans of the systems are possible. This dissertation contends that to deal with future uncertainties, flexibility in urban drainage systems is necessary. At present, profound insights about defining, measuring, and generating flexible urban drainage systems do not exist. This research systematically approaches these issues. First, a clear definition of flexibility and an approach for the measurement and optimization of flexibility is operationalized. Based on the generic definitions of flexibility used in other disciplines, a definition tailored for urban drainage systems is generated. As such, flexibility in sustainable urban drainage systems is defined as `the ability of urban drainage systems to use their active capacity to act and respond to relevant alterations during operation in a performance-efficient, timely, and cost-effective way'. Next, a method for measuring flexibility is provided based on the developed definition of flexibility including the metrics, 'range of change', 'life-cycle performance' and 'effort of change'. These metrics are integrated into a framework for the measurement of flexibility based on a comparison of performance and effort in different alternative solution with respect to different future states. In addition the metrics are the core components for optimizing flexible design of urban drainage system. The measurement method is successfully applied in two case studies in Tuttle Hill, UK and Hamburg-Wilhelmsburg, Germany. Using the developed definition and method for the measurement of flexibility, this dissertation illustrates that a transfer of the general theoretical background of flexibility to the field of urban drainage is possible. It is currently unclear how the flexible design of urban drainage systems can be executed. Based on a review, this research identifies nine potential principles of flexible design, described by the indicators of modularity, platform design, flexible elements, cost efficiency, decentralized design, real time control, low degree of specialization, scalability, and a combination of these principles. A case study of Hamburg-Boberg is then presented to analyze which of these principles of flexible design can be verified. For each alternative solution in the sample, the indicators for the different potential principles of flexible design as well as the flexibility provided by the design are calculated. Testing is done to determine if there is a significant correlation between the potential principles of flexible design and the measured flexibility using a chi-square-test and F-test. Two principles are verified with a high degree of confidence, 'platform design' and `flexible elements'. The `platform design' principle provides high flexibility, in which urban drainage system elements with high change costs are designed robustly with huge tolerance margins, whereas elements with low change costs are designed with flexibility options. The 'flexible elements' principle aims to include as many component elements as possible, which provides high individual flexibility in the design of the urban drainage system. These design principles and associated static indicators enable a quick screening of huge number alternative solutions and provide guidance for the development and optimization of flexible urban drainage system. Within the framework for optimization of flexibility, the design principles can help identify the most promising alternative solutions for the design of urban drainage systems. The optimization framework includes the following steps: identification of the required flexibility, generation of alternative solutions for the design of urban drainage systems, screening of the most promising alternative solutions, detailed measurement of flexibility provided by the alternative solutions; and selection of optimal solution. Hence out of a sample of different design approaches, the solutions with the highest flexibility could be identified. The successful application of flexible design in three case studies illustrates that the concept provides a suitable strategy for dealing with the challenges associated with future uncertainties. For urban drainage systems, flexible design guarantees high levels of performance in uncertain future states while reducing the effort required to adapt the system to changing future conditions. This study contends that flexibility allows for profound decision making for urban drainage design despite future uncertainties.

Both real-time virtualization and recursive virtualization are desirable properties of a virtual machine monitor (or hypervisor). Although the prospect for virtualization and even recursive virtualization has become better as the PC hardware becomes faster, the real-time systems community so far has not been able to reap much benefits. This is because no existing virtualization mechanism can properly support the stringent timing requirements needed by real-time systems. It is hard to do real-time virtualization, and it is even harder to do it recursively. In this dissertation, we propose a framework whereby the hypervisor is capable of running real-time guests and participating in recursive virtualization. Such a hypervisor is called a real-time hierarchical hypervisor. We first look at virtualization of abstract resource types from the real-time systems perspective. Unlike the previous work on recursive real-time partitioning that assumes fully-preemptable resources, we concentrate on other and often more practical types of scheduling constraints, especially the non-preemptive and limited-preemptive ones. Then we consider the current x86 architecture and explore the problems that need to be addressed for real-time recursive virtualization. We drill down on the problem that affects timing properties the most, namely, the recursive forwarding and delivery of interrupts, exceptions and intercepts. We choose the x86 architecture because it is popular and readily available, but it is by no means the only architecture of choice for real-time recursive virtualization. We conclude the research with an architecture-independent discussion on future possibilities in real-time recursive virtualization. / text

No description available.