Load testing remains the most integral part of testing and measuring the performance of Large Scale Software Systems (LSS). During the course of a load test, a system under test is closely monitored, resulting in an extremely large amount of logging data, e.g., Performance counters logs. The performance counter log captures run-time system properties such as CPU utilization, disk I/O, queues, and network traffic. Such information is of vital interest to performance analysts. The information helps them to observe the system’s behavior under load by comparing it against the documented behavior of a system or with expected behavior. In practice, for LSS, it is impossible for an analyst to skim through the large amount of performance counters to find the required information. Instead, analysts often use ‘rules of thumb’. In a LSS, there is no single person with complete system knowledge. In this thesis, we present methodologies to help performance analysts to 1) more effectively compare load tests to detect performance deviations, which may, lead to Service Level Agreement (SLA) violations and 2) provide them with a smaller and manageable set of important performance counters to assist in the root cause analysis of the detected deviations. We demonstrate our methodologies through case studies based on load test data obtained from both a large scale industrial system and an open source benchmark system. Our proposed methodologies can provide up to 89% reduction in the set of performance counters while detecting performance deviations with few false positives (i.e., 95% average precision). / Thesis (Ph.D, Computing) -- Queen's University, 2013-08-28 23:04:58.774

Orientador: David de Carvalho / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Civil, Arquitetura e Urbanismo / Made available in DSpace on 2018-08-21T12:02:47Z (GMT). No. of bitstreams: 1 Kassouf_Roberto_M.pdf: 4587745 bytes, checksum: 67d8f7ab10c6a6a1d4b0e8a2abdfa71e (MD5) Previous issue date: 2012 / Resumo: O problema de tubulões carregados horizontalmente no topo é freqüente em obras de construção civil e normalmente estes esforços horizontais atuam concomitantemente aos esforços verticais e momentos fletores onde os casos mais freqüentes são as ações dos ventos, torres de transmissão e para energia eólica, obras de arte como pontes e viadutos, muros de arrimo com o empuxo do solo, escavações para subsolo e edificações prediais, havendo uma reduzida bibliografia sobre os parâmetros a serem adotados em projeto. Uma hipótese mais critica é a de tubulões carregados horizontalmente em solos de alta porosidade, não saturados e colapsíveis. Os objetivos desta dissertação são: (1) proporcionar subsídios para o projeto de fundações com carregamento horizontal para tubulões, chamando a atenção para o danoso efeito de uma inundação de um solo colapsível; (2) fornecer valores do coeficiente de reação horizontal (nh), para uma argila siltosa arenosa de alta porosidade, solo de diabásio, comum na região de Campinas, para condição natural de umidade e solo pré-inundado; (3) comparar os valores obtidos com valores de (nh) com os propostos na bibliografia. Para a analise deste problema foi executado um tubulão em maciço de solo de formação de magmáticos básicos, ocorrendo rochas intrusivas básicas (Diabásios) da Formação Serra Geral, que faz parte do Grupo São Bento, característica do Campo Experimental de Fundações da UNICAMP/Campinas, com mais dois tubulões alinhados, estes servindo de reação ao sistema, todos executados a céu aberto e com base alargada, onde foram inseridos no eixo do tubulão a ser analisado, tubos para utilização de inclinomêtro e strain gages. As análises basearam-se nos resultados obtidos nas curvas de cargas versus deslocamento horizontais resultantes no topo e em profundidade através da utilização do inclinômetro / Abstract: The problem of caissons loaded horizontally at the top is common in construction work and normally these horizontal forces act concurrently with the vertical forces and bending moments where the most frequent cases are the actions of the winds, in towers for electric transmission and for wind energy, works of art as bridges and viaducts, retaining walls, underground excavations for buildings and there is a small literature on the parameters to be adopted in the project. One hypothesis, the most critical, is of caissons loaded horizontally in soils of high porosity, unsaturated and collapsible. The objectives of this dissertation are: (1) provide grants for the design of foundations with horizontal loading for caissons, calling attention to the damaging effect of a flood of a collapsible soil, (2) provide values of the coefficient of horizontal reaction (nh) to a sandy silty clay of high porosity, diabase soil, common in the region of Campinas, in the natural moisture content condition and in the pre-inundated soil condition, (3) compare the values obtained with the value (nh) with those proposed in the literature. For the analysis of this problem was carried out an caisson in a soil constituted by magmatic basic intrusive rocks (diabases) of the Serra Geral Formation, which is part of the São Bento Group, characteristic of the Experimental Foundations of Campinas / Campinas, aligned with two other caissons, serving as a reaction to these system, all running in the open and broad based, which were inserted into the pipe to be analyzed on its axis tubes for use with strain gages and inclinometer. The analyzes were based on the results of the load versus displacement curves obtained in the caisson head and in caisson shaft in depth / Mestrado / Geotecnia / Mestre em Engenharia Civil

In Florida a number of highway bridges were retrofitted on their reinforced concrete (RC) girders with carbon-fiber reinforced polymers (CFRP) during the 1990's. Their conditions, after being in service for approximately 10 years, are of significant interest to the State's highway authority, as well as researchers in the region. This paper will evaluate if a load test on one of such bridges, which was retrofitted with CFRP at the girders in the splash-zone and thus was subjected to severe environmental conditions, is a feasible technique to evaluate the actual condition of the CFRP. A 3-dimensional Finite Element Model (FEM) was utilized to assess the load-deflection behavior of the bridge. An analytical study was used to evaluate the effective moment of inertia of the strengthened beams modeled on the FEM. The results indicate that the deflection change due to the amount of CFRP sheets assumed to be effective on the beam is insignificant. The paper also shows that it would not be feasible to estimate changes in the properties in the CFRP based only on deflection and strain measurements.

A prova de carga dinâmica é freqüentemente utilizada na determinação da capacidade de carga de fundações por estacas cravadas e há relatos de uso com sucesso deste ensaio em estacas escavadas e tubulões. Esta pesquisa envolveu a adaptação e execução da prova de carga dinâmica na placa de aço circular de diâmetro 0,80 m, com o objetivo de inferir a capacidade de carga do sistema placa-solo. Como a prova de carga dinâmica permite que seja feita a comprovação da carga durante a cravação, vislumbra-se também a viabilização desse tipo de ensaio em solos porosos por meio da cravação da placa o necessário para atingir a capacidade de carga exigida em um projeto. Os ensaios foram realizados no campo experimental da USP/São Carlos, com solo superficial não saturado, poroso e colapsível, por isso foram feitas medidas dos valores de sucção matricial, já que esta exerce forte influência sobre a capacidade de carga / The dynamic load test is frequently used in the determination of driven piles foundation bearing capacity. It is also reported the successful use of the test in bored piles and caissons. This research included the adaptation and execution of the dynamic load test in circular steel plate, to verify the bearing capacity of footing foundations. Since the dynamic load tests allow the bearing capacity to be verified during the construction, it is also glimpsed the possibility of using this test in porous soils by penetrating the plate into the soil until the expected bearing capacity is achieved. The tests were held at the USP/São Carlos experimental foundation camp, which has superficial unsaturated, porous and colapsive soil, so matric suction's measurement were made, since it has strong influence on the bearing capacity

Soils, like several other materials, exhibit strong time-dependent behaviour which can be evidenced in terms of creep or strain-rate effects. The degree of this rheological behaviour varies with the type of soil, its structure, and with the stress history. This effect is exacerbated in pile load testing where the procedure duration tends to be shortened under increasing time pressures. The modelling needed to interpret the results therefore becomes more and more complex, including soil viscosity, wave radiation into the soil and other significant phenomena. The objective of the research reported herein is to refine the rheological parameters characterizing the influence of the loading rate within the framework of a relevant pile/soil interaction model fed with dynamic measurements acquired during pile Dynamic Load Tests (DLTs). The final goal is to predict and simulate the quasi-static pile load settlement curve. The pile/soil interaction system is described by a non-linear mass/spring/dashpot system supposed to represent the pile and the soil, with constitutive relationships existing within and between them. These relationships account for the static and the dynamic or rheologic behaviour. A back-analysis process based on a matching procedure between measured and computed quantities allows one to characterize the pile/soil interaction in terms of constitutive and rheologic parameters based on the dynamic measurements. After optimisation of the matching procedure, the parameters obtained are used to simulate the “static” load-settlement curve. The matching procedure is based on an automatic and stochastic parameter perturbation analysis. Since the parameters influence the system response with a relative weight, they are sorted in order to optimise all the parameters by successively retrieving the most influential ones and working on the remaining ones. The back-analysis performed on real dynamic measurements in this research leads to an improved pile/soil interaction model. The slippage between pile and soil along the pile shaft must be explicitly taken into account. This refinement increases the number of degrees of freedom needed to describe the pile/soil system but brings deeper insight into the behaviour of an interfacing zone of limited thickness surrounding the pile shaft.

A prova de carga dinâmica é freqüentemente utilizada na determinação da capacidade de carga de fundações por estacas cravadas e há relatos de uso com sucesso deste ensaio em estacas escavadas e tubulões. Esta pesquisa envolveu a adaptação e execução da prova de carga dinâmica na placa de aço circular de diâmetro 0,80 m, com o objetivo de inferir a capacidade de carga do sistema placa-solo. Como a prova de carga dinâmica permite que seja feita a comprovação da carga durante a cravação, vislumbra-se também a viabilização desse tipo de ensaio em solos porosos por meio da cravação da placa o necessário para atingir a capacidade de carga exigida em um projeto. Os ensaios foram realizados no campo experimental da USP/São Carlos, com solo superficial não saturado, poroso e colapsível, por isso foram feitas medidas dos valores de sucção matricial, já que esta exerce forte influência sobre a capacidade de carga / The dynamic load test is frequently used in the determination of driven piles foundation bearing capacity. It is also reported the successful use of the test in bored piles and caissons. This research included the adaptation and execution of the dynamic load test in circular steel plate, to verify the bearing capacity of footing foundations. Since the dynamic load tests allow the bearing capacity to be verified during the construction, it is also glimpsed the possibility of using this test in porous soils by penetrating the plate into the soil until the expected bearing capacity is achieved. The tests were held at the USP/São Carlos experimental foundation camp, which has superficial unsaturated, porous and colapsive soil, so matric suction's measurement were made, since it has strong influence on the bearing capacity

Foram construídas bases injetadas próximas à superfície e estacas escavadas com base injetada no campo experimental de fundações do departamento de geotecnia da USP/São Carlos. As injeções de caldas de cimento portland foram realizadas como procedimento de melhoria do solo de apoio. Provas de carga estática em estruturas com e sem injeção determinaram a eficácia da solução de bases injetadas. Simultaneamente às provas de carga, houve monitoração da sucção matricial atuante no solo, dando continuidade ao extenso programa de pesquisa realizado no Campo Experimental de Fundações da USP/São Carlos no tema da colapsibilidade. / Injected elements close to surface and toe injected bored piles were built at USP/São Carlos experimental foundation field. Under the foundations, portland cement grout was injected in order to improve soil and foundation bearing capacity. Load tests measured and determined the efficacy of the solution. Along with the load tests, matric suction measurements were made, continuing the extensive research program in the USP/São Carlos experimental foundation field under the theme soil collapsibility.

Foram construídas bases injetadas próximas à superfície e estacas escavadas com base injetada no campo experimental de fundações do departamento de geotecnia da USP/São Carlos. As injeções de caldas de cimento portland foram realizadas como procedimento de melhoria do solo de apoio. Provas de carga estática em estruturas com e sem injeção determinaram a eficácia da solução de bases injetadas. Simultaneamente às provas de carga, houve monitoração da sucção matricial atuante no solo, dando continuidade ao extenso programa de pesquisa realizado no Campo Experimental de Fundações da USP/São Carlos no tema da colapsibilidade. / Injected elements close to surface and toe injected bored piles were built at USP/São Carlos experimental foundation field. Under the foundations, portland cement grout was injected in order to improve soil and foundation bearing capacity. Load tests measured and determined the efficacy of the solution. Along with the load tests, matric suction measurements were made, continuing the extensive research program in the USP/São Carlos experimental foundation field under the theme soil collapsibility.

Adjacent box beam bridges are economical bridge systems for accelerated bridge construction. The box beams are constructed at precast plants and are traditionally connected by a shear key filled with grout. This system is ideal for short spans with low clearance restrictions. However, due to the grout deteriorating and debonding from the precast concrete in the shear key, reflective cracking propogates through the deck allowing water and chemicals to leak down into the joints. This can lead to the prestressing steel inside the precast member and the transverse tie steel corroding. This necessitates the bridge being rehabilitated or replaced which shortens the life-span of the bridge system and negates the economical value it had to begin with. This research project aimed to design a rehabilitation plan for an adjacent box beam bridge with deteriorated joints using Very High Performance Concrete (VHPC). VHPC was chosen as an economical alternative to the proprietary Ultra High Performance Concrete (UHPC) and extensive material tests were performed. The results of the material testing of VHPC and grout revealed that VHPC had higher compressive and tensile strengths, a higher modulus of elasticity, gained strength faster, bonded better to precast concrete, was more durable over time, and shrank less than conventional grout. The results of this research project were applied to rehabilitate the Buffalo Branch Bridge and further testing will be completed to determine the effectiveness of the rehabilitation. / Master of Science

No description available.