Making computer systems more energy efficient while obtaining the maximum performance possible is key for future developments in engineering, medicine, entertainment, etc. However it has become a difficult task due to the increasing complexity of hardware and software, and their interactions. For example, developers have to deal with deep, multi-level cache hierarchies on modern CPUs, and keep busy thousands of cores in GPUs, which makes the programming process more difficult. To simplify this task, new abstractions and programming models are becoming popular. Their goal is to make applications more scalable and efficient, while still providing the flexibility and portability of old, widely adopted models. One example of this is task-based programming, where simple independent tasks (functions) are delegated to a runtime system which orchestrates their execution. This approach has been successful because the runtime can automatically distribute work across hardware cores and has the potential to minimize data movement and placement (e.g., being aware of the cache hierarchy). To build better runtime systems, it is crucial to understand bottlenecks in the performance of current and future multicore systems. In this thesis, we provide fast, accurate and mathematically-sound models and techniques to understand the execution of task-based applications concerning three key aspects: memory behavior (data locality), scheduling, and performance. With these methods, we lay the groundwork for improving runtime system, providing insight into the interplay between the schedule's behavior, data reuse through the cache hierarchy, and the resulting performance. / UPMARC

No description available.

This study aims to investigate what can be done to make internet users more aware of what information they share with service providers, as well as to serve as a pilot test for two prototypes as a part of a EU funded project called CREDENTIAL. For these tests 12 participants were recruited. They were divided into two groups, where one group used a tablet for the tests and the other group used a computer. The test participants are introduced to the tests under the pretense that they are going to rate photographs for a given category, when in reality the purpose is to study the participants interaction with the dialogues requesting personal information they have to handle in order to receive the photos to rate. The study shows that test participant rate both prototypes similarly in terms of usability, but for actually remembering what information they shared the prototype using drag and drop functionality proved superior. As for the study as a pilot test, several issues with both prototypes have been identified and possible solutions to these issues are presented.

Designing an application from top to bottom is a challenge for any software architect. Designing an application to be deployed in the cloud adds extra complexity and a variety of questions to the task. One of these questions is how to deploy an application? The most popular choices at this time are either Docker containers or serverless functions. This report presents a comparison between the two deployment methods based on cost and performance. The comparison did not yield a conclusive winner but it did offer some key pointers to help with the decision. Docker containers offer a standardized deployment method for a low price and with good performance. Before choosing Docker the intended market needs to be evaluated given that for each region Docker needs to serve, the price also increases. Serverless functions offer auto-scaling and easy global deployments but suffer from high complexity, slower performance, and an uncertain monthly price tag.

In this project, we use a computer to sense, extract features from the signal,synthesize and control a graphical ball. Our aim is to control children respirationby software, displaying a small ball in a screen. When the child breathes on themicrophone, the ball moves in the display so the doctor will have time/possibilityto listen to the lung with a stethoscope.We used Action Script3.0 to write our project codes. The resulting software isexecuted in (Adobe Inc.trademark software environment) Flash.

We have investigated the page coloring technique bank partitioning and if it can be applied on commercial hardware platforms to reduce execution time jitter for specific tasks. We have also investigated how to alter execution times using bank partitioning. Unpredictable latency created by execution time jitter is a problem in real-time computing on commercial hardware platforms. We have run experiments that try to prove that the bank partitioning method we use alters the execution time and that thrashing occurs in the main memory if we run multiple instances of a workload. We receive significant changes in execution times when using bank partitioning and we can determine that thrashing occurs. However, due to the lack of the ability to measure the hardware performance counter for row buffer misses, we cannot determine if thrashing occurs in the main memory level. Since we cannot determine when, or if thrashing occurs in the main memory we find that we cannot reduce execution time jitter on the two systems that we have tested using bank partitioning on. We also find that execution times of specific tasks can be altered by reducing the number of bank bins associated with the specific task. The execution time of the task is increased if we reduce the number of bins associated with it.

No description available.

Möjligheten att fjärrstyra flygledartorn kommer att ställa högre krav på flygledares koncentrations- och simultanförmåga runt om i Sverige. Det är viktigt att åtgärder tas för att förhindra att en trött flygledare begår ett misstag och det är just detta som det framtagna systemet försöker att förhindra. Med hjälp av sensorer och modeller kan systemet bestämma flygledarens trötthet, stressnivå, uppmärksamhet och nuvarande arbetsuppgift. Alla värden presenteras i ett enkelt grafiskt gränssnitt. Tillsammans med resultaten för flygledarens hälsa presenteras även all sensordata i gränssnittet. Systemet är främst uppbyggt av två olika ramverk: Apache NiFi och Apache Spark. Vad de båda ramverken har gemensamt är att de har funktionalitet för att bygga kluster, vilket betyder att endast antalet noder sätter gränsen för hur många flygledare som kan vara uppkopplade samtidigt. Denna prototyp har inte all funktionalitet på plats för att behandla flera flygledare. Grunden är däremot lagd för att enkelt kunna implementera ytterligare funktionalitet och i slutändan ha flera flygledare uppkopplade samtidigt. Systemet öppnar upp möjligheter till att fördela arbetet på de flygledare som är mest fokuserade och kan därför bidra till att öka flygsäkerheten.

Electronic devices are widely used in today’s hospitals and the possibilities they offer are increasing every day. The devices are often embedded systems, running outdated operating systems and need to have a high uptime which makes them vulnerable to malicious software (malware). This thesis examines the ways malware can propagate through the Universal Serial Bus (USB) with the help of social engineering. Valuable assets are defined and different threat scenarios to these assets are presented. Lastly, the different scenarios are evaluated based on which assets they impact and how to effective mitigate the threats they present. Short- and long-term mitigations are presented to secure the devices in a broader view.

Cloud computing promises a new era of service delivery and deployment in such a way that every person can access any kind of services like storage, application, operating system and so on from anywhere any time using any device having internet connection. Cloud computing opens new possibilities approaching sustainable solutions to deploy and advance their services upon that platform. Sustainability of Cloud computing is to be addressed in terms of environmental and economic effects.In this thesis we explore the energy efficient approaches inside data centres from the site and IT infrastructure perspective incorporating Cloud networking from the access network technologies and network equipment point of view to give a comprehensive prospect toward achieving energy efficiency of Cloud computing. Traditional and Cloud data centres would by compared to figure out which one is more recommended to be deployed. Virtualization as heart of energy efficient Cloud computing that can integrates some technologies like consolidation and resource utilization has been introduced to prepare a background for implementation part. Finally approaches for Cloud computing data centres at operating system and especially data centre level are presented and Green Cloud architecture as the most suitable green approach is described in details. In the experiment segment we modelled and simulated Facebook and studied the behaviour in terms of cost and performance and energy consumption to reach a most appropriate solution.