Thesis (M.S.)--University of Missouri-Columbia, 2005. / The entire dissertation/thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file (which also appears in the research.pdf); a non-technical general description, or public abstract, appears in the public.pdf file. Title from title screen of research.pdf file viewed on (January 24, 2007) Includes bibliographical references.

There is a movement underway to eliminate the practice of tipping restaurant servers that is gaining momentum (Goldberger, 2015). In lieu of gratuities, restaurants are simply raising menu prices or assessing a service charge and paying servers a fixed hourly wage (Kummer, 2016). Before restaurateurs can adopt such a strategy, they need to thoroughly understand the factors that affect tipping behavior in order to develop meaningful fixed wage rates that do not diminish service levels or employee morale. The first step in this process is a better understanding of the determinants of tipping rates. The existing research has identified many factors that influence tipping rates. Some factors are outside of the server’s control such as the server’s and the customer’s race, the size of the bill, the size of the dining party, and whether or not alcohol was consumed. Other factors are within the servers control such as the quality of the service, whether they squatted at the table, or wrote a smiley face on the guest check. Most of this research, however, is based on empirical analysis using small samples sizes and/or questionnaires that may not reflect actual behavior, or data from interviews based on what consumers say they did. The contribution of this research is that I identified two additional significant determinants of tipping rates, sales tax rates and discount rates, which have not previously been studied. This research also extends the previous research related to the impact of bill size and dining party size but with a significantly larger sample. The study presented herein includes an analysis over 75 million guest checks from 43 brand-name restaurants across 1,202 locations over three years to understand precisely how customers behaved. The data were analyzed using a two-way fixed effects, ordinary least squares model to limit time and spatial controls to a single national time trend with state fixed effects. The state fixed effects control for differences across states that are fixed over time and quarterly fixed effects will control for factors that impact tipping rates equally across all states in a given quarter. Robust standard errors were clustered for the 43 restaurant brands and twelve quarters of time to account for temporal serial correlation in the error terms within the locations. The analysis revealed that there is a positive relationship between sales tax rates and tipping rates and an inverse relationship between discount rates and tipping rates. This essentially implies that consumers are tipping on the post-tax bill net of any discounts. The study also confirms the results of certain prior research in that both party size and bill size are inversely related to tipping rates up to a certain point.

Server selection is an important problem of cloud computing in which cloud service providers direct user demands to servers in one of the multiple data centers located in different geographical locations. The existing solutions usually assume homogeneity of cloud services (i.e., all users request the same type of service) and handle user demands in an individual basis which incurs high computational overhead. In this study, we propose a new and effective server selection scheme in which diversities of cloud services are taken into account. We focus on a specific cloud service, i.e., online video service, and assume that different videos have different bandwidth requirements. We group users into clusters and handle user demands on a cluster basis for faster and more efficient process. Firstly, we assume that user demands and bandwidth capacities of servers are given in the data centers, our problem is to assign the user demands to the servers under the bandwidth constraint, such that the overall latency (measured by the network distance) between the user clusters and the selected servers is minimized. We design a server selection system and formulate this problem as a linear programming formulation which can be solved by existing techniques. The system periodically executes our scheme and computes an optimal solution for server selection. User demands are assigned to the servers according to the optimal solution and the minimum overall latency can be achieved. The simulation results show that our scheme is significantly better than the random algorithm and the YouTube server selection strategy. Based on the first part, we take the storage capacities of servers constraint into consideration. In the second part, our new problem is to assign the user demands to the servers under the bandwidth and storage constraint, such that the function of overall latency (measured by the network distance) between the user clusters and the selected servers and standard deviation of traffic load of every server in the system is minimized. We design a server selection system and formulate this problem which can be solved by existing techniques. User demands are assigned to the servers according to the optimal solution and the two goals (minimum overall latency and the most balanced traffic load) can be achieved. The simulation results show the influence of different weights of these two goals on the user demands assigning.

No description available.

Embedded/real-time systems are becoming ubiquitous in today's world and their pervasive nature is increasing with the advent of cyber-physical systems. Providing temporal guarantees is paramount in such systems. Most of the normal operation in real-time systems is modelled using periodic tasks. Event-driven behaviour is modelled using aperiodic jobs. To ensure an acceptable Quality of Service for aperiodic jobs without jeopardizing safety of periodic tasks, aperiodic servers were introduced [2], [3]. Aperiodic servers are used to reserve a quota for the execution of aperiodic jobs. However, they do not take into account, cache-related delays that the execution of aperiodic jobs could impose on periodic tasks, thereby making their use in systems with caches unsafe. In this thesis, we introduce Cache Related Delay Servers to solve this problem. Statically, every periodic task's worst-case execution time includes a pre-determined delay quota for delay caused by aperiodic jobs. During system operation, the aperiodic server is allowed to execute only if periodic jobs that may be affected by it have sufficient delay quota to accommodate its execution. Otherwise, the priority of the aperiodic server is temporarily decreased to the level of the lowest-priority periodic job with insufficient quota, thereby ensuring safe execution of periodic tasks.

published_or_final_version / Computer Science and Information Systems / Master / Master of Philosophy

O surgimento de novos serviços e aplicações baseados na Web tem provocado um aumento desenfreado na quantidade de usuários da World Wide Web que, por sua vez, se torna cada vez mais popular no mundo dos negócios. Sites de e-commerce, que demandam grande tráfego de requisições, têm adotado sistemas de servidores Web distribuídos, como a arquitetura Web Cluster. Isso se deve ao fato de enfrentarem frequentemente situações de sobrecarga, durante as quais podem deixar de atender requisições de transação (com grande probabilidade de gerar renda) por conta do aumento na demanda de requisições de navegação (geram renda apenas de forma indireta). A utilização ineficiente de recursos pode comprometer o desempenho do sistema e é nesse contexto que este trabalho se insere. Neste trabalho foi desenvolvido um modelo de Servidor Web para E-Commerce (SWE-C),validado por meio de um modelo de simulação e uma carga sintética gerada a partir de um modelo desenvolvido com os principais tipos de requisições que caracterizam um site de e-commerce. Foram realizadas simulações no sistema com diversas combinações de algoritmos de escalonamento e disciplinas de atendimentos para filas, dentre as quais de destaca uma nova disciplina que utiliza um mecanismo de prioridades orientado ao consumo de CPU proposto neste trabalho. O objetivo é aumentar o throughput de requisições de transação e melhorar os tempos de resposta em situações de sobrecarga. Uma avaliação de desempenho foi realizado e constatou-se que o mecanismo de prioridades proposto é adequado às necessidades de um site de e-commerce. / The appearance of new services and applications based on the web has causing a wild increase in the amount of users of the World Wide Web that becomes more popular in the business world. E-commerce sites that demand great request traffic has adopted a distributed web servers system, such as the Web Cluster. This occurs because this sites frequently are on overloaded situations, leaving to serve request transactions (with large probability of generate income) due to increasing demand of navigation requests (generates income only indirectly). The low resource utilization can compromise the system performance. This work is inserted in this context. In this master thesis has been developed a Web server model for E-Commerce (WSE-C). It is validated with a simulation model using a synthetic workload characterized with the main types of equest identified from e-commerce sites. Several simulations were accomplished in the system, combining the scheduling algorithms, the queue attendances disciplines and the use of a new priority mechanism oriented to the CPU utilization. The aim of this work is increase the request throughput and to obtain a better response time on overload situation. A performance evaluation was conducted and shown that priority mechanisms is adequate to a e-commerce site.

In this thesis, three key issues that restrict networks from scaling up so as to be able to cope with the rapid increase in traffic are investigated and series of approaches are proposed and tested for overcoming them. Firstly, scalability limitations owing to the use of a broadcast mechanism in one collision domain are discussed. To address this matter, servers under software-defined network architectures for eliminating discovery messages (SSED) are designed in this thesis and a backbone of floodless packets in an SDN LAN network is introduced. SSED has an innovative mechanism for defining the relationship between the servers and SDN architecture. Experimental results, after constructing and applying an authentic testbed, verify that SSED has the ability to improve upon the scalability of the traditional mechanism in terms of the number of switches and hosts. This is achieved by removing broadcast packets from the data and control planes as well as offering a better response time. Secondly, the scalability restrictions from using routers and the default gateway mechanism are explained. In this thesis, multiple distributed subnets using SDN architecture and servers to eliminate router devices and the default gateway mechanism (MSSERD) are introduced, designed and implemented as the general backbone for scalable multiple LAN-based networks. MSSERD's proposed components handle address resolution protocol (ARP) discovery packets and general IP packets across different subnets. Moreover, a general view of the network is provided through a multi-subnets discovery protocol (MDP). A 23 computers testbed is built and the results verify that MSSERD scales up the number of subnets more than traditional approaches, enhances the efficiency significantly, especially with high load, improves performance 2.3 times over legacy mechanisms and substantially reduces complexity. Finally, most of the available distributed-based architectures for different domains are reviewed and the aggregation discovery mechanism analysed to establish their impact on network scalability. Subsequently, a general distributed-centralised architecture with open-level control plane (OLC) architecture and a dynamic discovery hierarchical protocol (DHP) is introduced to provide better scalability in an SDN network. OLC can scale up the network with high performance even during high traffic.

Este estudo buscou compreender o canal de marketing usado para distribuição de servidores de microinformática do tipo x86 que, segundo institutos de pesquisas como o IDC (2009), responderão pela maioria absoluta de todos os servidores vendidos no país. Dentro da categoria de servidores x86, foram explorados os que estão em fases de crescimento e maturidade no ciclo de vida, devido aos potenciais de venda e de agregação de serviços ao cliente final. Para tanto, foram entrevistados doze membros deste canal, incluindo o fabricante dos produtos, três atacadistas, quatro varejistas e quatro clientes finais. Estes foram questionados sobre suas funções no canal de marketing analisado, avaliação de capacitação técnica para venda e prestação de serviços ao cliente, além dos papéis dos clientes e suas expectativas quando ao serviço ao cliente prestado. Percebeu-se que deficiência na capacitação e falta de estruturação nos varejistas pode ser um fator crítico para o canal de marketing, podendo comprometer a imagem dos produtos vendidos, o entendimento das necessidades e a prestação de serviços ao cliente e, de certa forma, o resultado de todos os outros membros do canal. / The intent of this study was to understand the marketing channel used to distribute x86 servers which, according to research institutes such as IDC (2009), will account for the majority of the servers sold in Brazil. Within the x86 category only products in growth or maturing lifecycle were considered in the investigation. These were selected due to potential of sales and customer services aggregation to the final purchasing customers. To accomplish that, twelve members of this channel were interviewed, including the product manufacturer, three wholesalers, four retailers and four final customers. They were asked about their function in the studied marketing channel, about their technical capability to sell and to serve customers, in addition to understanding the customer roles and their expectations on the performed service. The results found that reseller deficiency in technical skills and poor structure definition may be a critical factor for the marketing channel. This possibly compromises the image of sold products, the understanding of customer needs, customer services delivered and, in a certain way, the results for the other channel members.

Not available