Offloading tasks to a network processor is one of the important ways to increase server performance. Hardware offloading of Transmission Control Protocol/Internet Protocol (TCP/IP) intensive tasks is known to significantly improve performance. When the entire application is considered for offloading, the impact on the server can be significant because it significantly reduces the load on the server. The goal of this thesis is to consider such a system with application-level offloading, rather than hardware offloading, and gauge its performance benefits. I am implementing this project on an Apache httpd server (running RedHat Linux), on a system that utilizes a co-located network processor system (IXP2855). The performance of the two implementations is measured using the SPECweb2005 benchmark, which is the accepted industry standard for evaluating Web server performance.

No description available.

No / This paper investigates into the performance and load management of web servers that are deployed in commercial websites. Such websites offer various services such as flight/hotel booking, online banking, stock trading, and product purchases among others. Customers are increasingly relying on these round-the-clock services which are easier and (generally) cheaper to order. However, such an increasing number of customers' requests makes a greater demand on the web servers. This leads to web servers' overload and the consequential provisioning of inadequate level of service. This paper addresses these issues and proposes an admission control scheme which is based on the class-based priority scheme that classifies customer's requests into different classes. The proposed scheme is formally specified using -calculus and is implemented as a Java-based prototype system. The prototype system is used to simulate the behaviour of commercial website servers and to evaluate their performance in terms of response time, throughput, arrival rate, and the percentage of dropped requests. Experimental results demonstrate that the proposed scheme significantly improves the performance of high priority requests but without causing adverse effects on low priority requests.

Safe and efficient traffic control remains a challenging task with the continued increase in the number of vehicles, especially in urban areas. This manuscript focuses on traffic control at intersections, since urban roads with closely spaced intersections are often prone to queue spillbacks, which disrupt traffic flows across the entire network and increase congestion. While various intelligent traffic control solutions exist for autonomous systems, they are not applicable to or ineffective against human-operated vehicles or mixed traffic. On the other hand, existing approaches to manage intersections with human-operated vehicles, cannot adequately adjust to dynamic traffic conditions. This manuscript presents a technology-agnostic adaptive real-time server based approach to dynamically determine signal timings at an intersection based on changing traffic conditions and queue lengths (i.e., wait times) to minimize, if not eliminate, spillbacks without unnecessarily increasing delays associated with intersection crossings. We also provide timeliness guarantee bounds by analyzing the travel time delays, hence making our approach more dependable and predictable. The proposed approach was validated in simulations and on a realistic hardware testbed with robots mimicking human driving behaviors. Compared to the pre-timed traffic control and an adaptive scheduling based traffic control, our algorithm is able to avoid spillbacks under highly dynamic traffic conditions and improve the average crossing delay in most cases by 10--50 %. / Master of Science / Safe and efficient traffic control remains a challenging task with the continued increase in the number of vehicles, especially in urban areas. This manuscript focuses on traffic control at intersections, since urban roads with closely spaced intersections are often prone to congestion that blocks other intersection upstream, which disrupt traffic flows across the entire network. While various intelligent traffic control solutions exist for autonomous systems, they are not applicable to or ineffective against human-operated vehicles or mixed traffic. On the other hand, existing approaches to manage intersections with human-operated vehicles, cannot adequately adjust to dynamic traffic conditions. This work presents a technologyagnostic adaptive approach to dynamically determine signal timings at an intersection based on changing traffic conditions and queue lengths (i.e., wait times) to minimize, if not eliminate, spillbacks without unnecessarily increasing delays associated with intersection crossings. We also provide theoretical bounds to guarantee the performance of our approach in terms of the travel delays that may incur on the vehicles in the system, hence making our approach more dependable and predictable. The proposed approach was validated in simulations and on a realistic hardware testbed which uses robots to mimic human driving behaviour in an urban environment. Comparisons with widely deployed and state-of-the-art traffic control techniques show that our approach is able to minimize spillbacks as well as improve on the average crossing delay in most cases.

No description available.

This thesis investigates TCP connection management mechanisms in order to understand the behaviour and improve the performance of Internet servers during overload conditions such as flash crowds. We study several alternatives for implementing TCP connection establishment, reviewing approaches taken by existing TCP stacks as well as proposing new mechanisms to improve server throughput and reduce client response times under overload. We implement some of these connection establishment mechanisms in the Linux TCP stack and evaluate their performance in a variety of environments. We also evaluate the cost of supporting half-closed connections at the server and assess the impact of an abortive release of connections by clients on the throughput of an overloaded server. Our evaluation demonstrates that connection establishment mechanisms that eliminate the TCP-level retransmission of connection attempts by clients increase server throughput by up to 40% and reduce client response times by two orders of magnitude. Connection termination mechanisms that preclude support for half-closed connections additionally improve server throughput by up to 18%.

This study examines the role of traditional computing performance measures when used in a client/server system (C/SS) environment. It also evaluates the effectiveness of traditional computing measures of mainframe systems for use in C/SS. The underlying problem was the lack of knowledge about how performance measures are aligned with key business goals and strategies. This research study has identified and evaluated client/server performance measurements' importance in establishing an effective performance evaluation system. More specifically, this research enables an organization to do the following: (1) compare the relative states of development or importance of performance measures, (2) identify performance measures with the highest priority for future development, (3) contrast the views of different organizations regarding the current or desired states of development or relative importance of these performance measures.

Master of Science / Department of Computing and Information Sciences / Daniel A. Andresen / Recently, the world has been trending toward the use of Smartphone. Today, almost each and every individual is using Smartphone for various purposes benefited by the availability of large number of applications. The memory on the SD (Secure Digital) memory card is going to be a constraint for the usage of the Smartphone for the user. Memory is used for storing large amounts of data, which include various files and important document. Besides having many applications to fill the free space, we hardly have an application that manages the free memory according to the user’s choice. In order to manage the free space on the SD card, we need an application to be developed. All the important files stored on the Android device cannot be retrieved if we lose the Android device. Targeting the problem of handling the memory issues, we developed an application that can be used to provide the security to the important documents and also store unnecessary files on the distributed file servers and retrieve them back on request.

Mobile edge computing (MEC) has been considered as a promising technology to handle computation-intensive and latency-sensitive tasks for mobile user equipments (UEs) in next-generation mobile networks. Mobile UEs can offload these tasks to nearby edge servers, which are typically deployed on base stations (BSs) that are equipped with computation resources. Thus, the task execution latency as well as the energy consumption of mobile devices can be reduced. Mobility management has played a fundamental role in MEC, which associates UEs with the appropriate BSs. In the existing handover decision-making process, the communication costs dominate. However, in edge scenario, the computation capacity constraints should also be considered. Due to user mobility, mobile UEs are nonuniformly distributed over time and space. Edge servers in hot-spot areas can be overloaded while others are underloaded. When edge servers are densely deployed, each UE may have multiple choices to offload its tasks. Instead, if edge servers are sparsely deployed, UEs may only have one option for task offloading. This aggravates the unbalanced workload of the deployed edge servers. Therefore, how to serve the dynamic hot-spot areas needs to be addressed in different edge server deployment scenarios. Considering these two scenarios discussed above, two problems are addressed in this thesis: 1) with densely deployed edge servers, for each mobile UE, how to choose the appropriate edge servers independently without full system information is inves- tigated, and 2) with sparsely deployed edge servers, how to serve dynamic hot-spot areas in an efficient and flexible way is emphasized. First, with BSs densely de- ployed in hot-spot areas, mobile UEs can offload their tasks to one of the available edge servers nearby. However, precise full system information such as the server workload can be hard to be synchronized in real time, which also introduces extra signaling overhead for mobility management decision-making. Thus, a user-centric reinforcement-learning-based mobility management scheme is proposed to handle sys- tem uncertainties. Each UE observes the task latency and automatically learns the optimal mobility management strategy through trial and feedback. Simulation results show that the proposed scheme manifests superiority in dealing with system uncer- tainties. When compared with the traditional received signal strength (RSS)-based handover scheme, the proposed scheme reduces the task execution latency by about 30%. Second, fixed edge servers that are sparsely deployed around mobile UEs are not flexible enough to deal with time-varying task offloading. Dispatching mobile servers is formulated as a variable-sized bin-packing problem with geographic constraints. A novel online unmanned aerial vehicle (UAV)-mounted edge server dispatching scheme is proposed to provide flexible mobile-to-mobile edge computing services. UAVs are dispatched to the appropriate hover locations by identifying the hot-spot areas sequen- tially. Theoretical analysis is provided with the worst-case performance guarantee. Extensive evaluations driven by real-world mobile requests show that, with a given task finish time, the mobile dispatching scheme can serve 59% more users on aver- age when compared with the fixed deployment. In addition, the server utilization reaches 98% during the daytime with intensive task requests. Utilizing both the fixed and mobile edge servers can satisfy even more UE demands with fewer UAVs to be dispatched and a better server utilization. To sum up, not only the communication condition but also the computation lim- itation have an impact on the edge server selection and mobility management in MEC. Moreover, dispatching mobile edge servers can be an effective and flexible way to supplement the fixed servers and deal with dynamic offloading requests. / Graduate

Este trabalho propõe mecanismos de diferenciação de serviços para servidores Web, visando a melhorar o desempenho desses sistemas quando são consideradas as características das requisições Web nas políticas de atendimento. Optou-se por adotar o contexto do comércio eletrônico para a realização das pesquisas, uma vez que esse ambiente é um dos mais impactados negativamente quando há um comportamento inadequado do servidor em situações de sobrecarga. Para isso, foi realizada uma investigação das características das requisições Web típicas do e-commerce, para que tais características pudessem ser usadas como diretrizes para os mecanismos e melhorar o desempenho dos servidores. Em seguida, foram propostos um modelo de carga de trabalho e um modelo de simulação para a realização dos experimentos. Com isso, foi possível avaliar os resultados obtidos com a inserção dos diversos mecanismos no Servidor Web com Diferenciação de Serviços (SWDS), um modelo de servidor cuja arquitetura o torna capaz de fornecer serviços diferenciados a seus usuários e aplicações. Foram propostos novos mecanismos de escalonamento de requisições bem como novos mecanismos de controle de admissão. Diversas simulações foram realizadas e os resultados obtidos mostram que a exploração das características das requisições Web, além de ser fundamental para um bom entendimento do comportamento do servidor, possibilita a melhoria de desempenho do sistema / This work proposes differentiated services mechanisms for Web servers, aiming at improving their performance when the features of Web requests are considered. The electronic commerce (e-commerce) context was adopted to develop the researches once this environment is one of the most negatively influenced when there is an inadequate behavior of the server under overload situations. Thus, it was realized an investigation on the features of ecommerce Web requests, so that these features could be used both as guidelines for the mechanisms and to improve the performance of the servers. Afterwards, a workload model and a simulation model were proposed to implement the experiments. Thus, it was possible to evaluate the results obtained with the insertion of several mechanisms in the Web Server with Differentiated Services (WSDS), a server model with an architecture that makes it capable of supplying differentiated services to its users and applications. New request scheduling mechanisms were proposed as well as new mechanisms for admission control. Several simulations were realized and the obtained results show that the exploration of the Web request features, besides being fundamental for a good understanding of the server behavior, makes possible to improve the system performance