Since the turn of the century, the proliferation of virtualization and cloud computing has led to an increase in data centres and consequently an increase in power consumption for computing. Today, approximately 2% of global energy consumption is attributed to data centres alone. As a result, optimizing power usage effectiveness in enterprise data centres has become a laudable goal and a critical requirement in IT operations all over the world. While a significant body of research exists to measure, monitor, and control the “greenness” level of hardware components, significant research is needed to relate hardware energy consumption to energy consumption stemming from (software) program execution. In this dissertation, we argue that the true energy cost of program execution must focus on the digital ecosystem within which a particular software program is executed. We investigate the interplay between energy consumption, task scheduling and execution decision making using dynamic runtime models of digital ecosystems based on the execution context of software. Single instances of software applications are no longer confined to a single device or machine. Instead software commonly interacts with resources and services outside of its own hardware unit. The scope of this interaction defines the application’s digital ecosystem. Smartphones interact with cloud resources; cloud resources include databases, specialized compute or storage clouds, specialized hardware and virtual machines (VMs). Combining processes of varying complexity with varying resource allocations produces different energy consumption levels. The challenge is to investigate the variability of software process orchestration based on a power consumption framework to accrue and optimize energy savings in digital ecosystems. The contributions of this dissertation include: i) an adaptive energy consumption framework; ii) self-adaptive energy management systems based on this framework; iii) deployment mechanisms for applications to use this framework; iv) models at runtime models for self-adaptive energy management systems. Our ultimate goal is to develop smart, self-adaptive, green computing techniques, such as adaptive job scheduling and resource provisioning, to reduce overall power consumption in data centres, on individual devices (e.g., mobile, desktop, laptop or server), and in digital ecosystems. / Graduate

With rapid adoption of cloud solutions across industries, energy consumed by server farms continues to rise. There are numerous approaches to reduce energy consumption in data centres, and one of the approaches is to use energy-aware policies, which focus on how servers should be operated in order to achieve energy saving and meet service level agreements (SLA). In this thesis, we focus on studying a single server model with dynamic voltage scaling (DVS), presenting a framework with explicit solutions to solve for performance metrics and energy consumption. Our framework is convenient and in- tuitive, one can easily identify expected response time and expected energy consumption for a given policy. In addition, we also provide insights on how the value of the faster service rate and the choice of when to use speed scaling impact energy consumption and performance metrics. / Thesis / Master of Computer Science (MCS)

This dissertation contains three studies examining aspects of energy use by data centers and other information and communication technology (ICT) infrastructure necessary to support the electronic services that now form such a pervasive aspect of daily life. The energy consumption of ICT in general and data centers in particular has been of growing interest to both industry and the public, with continued calls for increased efficiency and greater focus on environmental impacts. The first study examines the metrics used to assess data center energy performance and finds that power usage effectiveness (PUE), the de facto industry standard, only accounts for one of four critical aspects of data center energy performance. PUE measures the overhead of the facility infrastructure but does not consider the efficiency of the IT equipment, its utilization, or the emissions profile of the power source. As a result, PUE corresponds poorly with energy and carbon efficiency, as demonstrated using a small set of empirical data center energy use measurements. The second study lays out a taxonomy of indirect energy impacts to help assess whether ICT’s direct energy consumption is offset by its energy benefits, and concludes that ICT likely has a large potential net energy benefit, but that there is no consensus on the sign or magnitude of actual savings, which are largely dependent upon implementation details. The third study estimates the potential of dynamic load shifting in a content distribution network to reduce both private costs and emissions-related externalities associated with electricity consumption. Utilizing variable marginal retail prices based on wholesale electricity markets and marginal damages estimated from emissions data in a cost-minimization model, the analysis finds that load shifting can either reduce data center power bills by approximately 25%–33% or avoid 30%–40% of public damages, while a range of joint cost minimization strategies enables simultaneous reduction of both private and public costs. The vast majority of these savings can be achieved even under existing bandwidth and network distance constraints, although current industry trends towards virtualization, energy efficiency, and green powermay make load shifting less appealing.

This study aims to investigate the main areas of Green IT and to determine which areas of Green IT have been widely adopted and implemented in our case studies. The purpose of the proposed study is also to identify, describe and analyze underlying factors that are perceived to be important to the adoption and implementation of Green IT. Finally, this research attempts to examine which factors are more important in our two case studies, Västerås City Stad and Mälardalen University.

Over the past few years, the energy consumption of Information and Communication Technologies (ICT) has become a major issue. Nowadays, ICT accounts for 2% of the global CO2 emissions, an amount similar to that produced by the aviation industry. Large-scale distributed systems (e.g. Grids, Clouds and high-performance networks) are often heavy electricity consumers because -- for high-availability requirements -- their resources are always powered on even when they are not in use. Reservation-based systems guarantee quality of service, allow for respect of user constraints and enable fine-grained resource management. For these reasons, we propose an energy-efficient reservation framework to reduce the electric consumption of distributed systems and dedicated networks. The framework, called ERIDIS, is adapted to three different systems: data centers and grids, cloud environments and dedicated wired networks. By validating each derived infrastructure, we show that significant amounts of energy can be saved using ERIDIS in current and future large-scale distributed systems.

With the resource constrained nature of mobile devices and the resource abundant offerings of the cloud, several promising optimisation techniques have been proposed by the green computing research community. Prominent techniques and unique methods have been developed to offload resource/computation intensive tasks from mobile devices to the cloud. Most of the existing offloading techniques can only be applied to legacy mobile applications as they are motivated by existing systems. Consequently, they are realised with custom runtimes which incur overhead on the application. Moreover, existing approaches which can be applied to the software development phase, are difficult to implement (based on manual process) and also fall short of overall (mobile to cloud) efficiency in software qualityattributes or awareness of full-tier (mobile to cloud) implications. To address the above issues, the thesis proposes a model-driven architecturefor integration of software quality with green optimisation in Mobile Cloud Applications (MCAs), abbreviated as Mango architecture. The core aim of the architecture is to present an approach which easily integrates software quality attributes (SQAs) with the green optimisation objective of Mobile Cloud Computing (MCC). Also, as MCA is an application domain which spans through the mobile and cloud tiers; the Mango architecture, therefore, takesinto account the specification of SQAs across the mobile and cloud tiers, for overall efficiency. Furthermore, as a model-driven architecture, models can be built for computation intensive tasks and their SQAs, which in turn drives the development – for development efficiency. Thus, a modelling framework (called Mosaic) and a full-tier test framework (called Beftigre) were proposed to automate the architecture derivation and demonstrate the efficiency of Mango approach. By use of real world scenarios/applications, Mango has been demonstrated to enhance the MCA development process while achieving overall efficiency in terms of SQAs (including mobile performance and energy usage compared to existing counterparts).

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / A crescente utilizaÃÃo de dispositivos mÃveis, redes sem fio e aplicaÃÃes mÃveis evidencia a importÃncia da garantia de seguranÃa da informaÃÃo. Esta preocupaÃÃo surge devido aos riscos envolvidos no trÃfego de informaÃÃes sensÃveis por meio sem fio, uma vez que o meio nÃo limita os riscos de ataques, tal como nas redes convencionais. Adicionalmente, a tendÃncia no uso de prÃticas sustentÃveis defendidas pela ComputaÃÃo Verde impÃe a necessidade de concepÃÃo de aplicaÃÃes flexÃveis que busquem a reduÃÃo do consumo de recursos, como o de energia. Assim, mecanismos para o provimento de confidencialidade de informaÃÃes que trafegam por meio sem fio devem considerar a alocaÃÃo eficiente de recursos computacionais. Esta à uma questÃo chave a ser considerada no momento da concepÃÃo de aplicaÃÃes mÃveis seguras. Portanto, os mecanismos de proteÃÃo devem balancear o nÃvel de seguranÃa requerido de acordo com o consumo de recursos alocados para provÃ-lo. O emprego de informaÃÃes que caracterizam a situaÃÃo corrente (contexto) pode auxiliar nessa tarefa. Assim, a utilizaÃÃo de proteÃÃo adequada aos requisitos de seguranÃa das aplicaÃÃes e combinada com o contexto pode identificar situaÃÃes nas quais serà necessÃrio aumentar ou diminuir o nÃvel de seguranÃa, de forma a diminuir o consumo de recursos do dispositivo. Esse trabalho propÃe, portanto, um Mecanismo de SeguranÃa com AdaptaÃÃo DinÃmica (MeSAD), com foco na confidencialidade, capaz de adaptar o nÃvel de seguranÃa de acordo com o contexto e reduzir o consumo de recursos dos dispositivos mÃveis. O objetivo principal consiste em encontrar o ponto de equilÃbrio no tradeoff entre nÃvel de seguranÃa e consumo de recursos. A fim de atingir este objetivo, este trabalho apresenta tambÃm uma ferramenta de suporte à utilizaÃÃo do MeSAD durante o desenvolvimento de aplicaÃÃes mÃveis, alÃm de possibilitar a realizaÃÃo de avaliaÃÃes sobre o desempenho dos algoritmos criptogrÃficos que sÃo utilizados nos diferentes dispositivos. / The increasing use of mobile devices, wireless networks and mobile applications highlights the importance of ensuring information security. This concern arises because of the risks involved in traffic sensitive information via wireless, since it does not limit the risk of attacks, as in conventional networks. Additionally, the trend in the use of sustainable practices advocated by the Green Computing imposes the need for designing flexible applications that seek to reduce consumption of resources such as energy. Thus, mechanisms for providing confidentiality of information passing over the wireless medium should consider the efficient allocation of computing resources. This is a key issue to be considered when designing secure mobile applications. Therefore, the protection mechanisms should balance the security level required in accordance with the consumption of resources allocated to provide it. The use of information that characterizes the current situation (context) can assist in this task. Thus, the use of appropriate protective security requirements of applications and combined with the context can identify situations where you need to raise or lower the security level in order to reduce the resource consumption of the device. This work proposes a Security Mechanism Dynamic Adaptation (MeSAD), focusing on confidentiality, able to adapt the level of security according to the context and reduce the resource consumption of mobile devices. The main objective is to find the balance point in the tradeoff between the level of security and resource consumption. In order to achieve this goal, this paper presents a tool to support the use of MeSAD during the development of mobile applications, and enable the assessments on the performance of cryptographic algorithms that are used in different devices.

The interest in machine learning algorithms is increasing, in parallel with the advancements in hardware and software required to mine large-scale datasets. Machine learning algorithms account for a significant amount of energy consumed in data centers, which impacts the global energy consumption. However, machine learning algorithms are optimized towards predictive performance and scalability. Algorithms with low energy consumption are necessary for embedded systems and other resource constrained devices; and desirable for platforms that require many computations, such as data centers. Data stream mining investigates how to process potentially infinite streams of data without the need to store all the data. This ability is particularly useful for companies that are generating data at a high rate, such as social networks. This thesis investigates algorithms in the data stream mining domain from an energy efficiency perspective. The thesis comprises of two parts. The first part explores how to extract and analyze data from Twitter, with a pilot study that investigates a correlation between hashtags and followers. The second and main part investigates how energy is consumed and optimized in an online learning algorithm, suitable for data stream mining tasks. The second part of the thesis focuses on analyzing, understanding, and reformulating the Very Fast Decision Tree (VFDT) algorithm, the original Hoeffding tree algorithm, into an energy efficient version. It presents three key contributions. First, it shows how energy varies in the VFDT from a high-level view by tuning different parameters. Second, it presents a methodology to identify energy bottlenecks in machine learning algorithms, by portraying the functions of the VFDT that consume the largest amount of energy. Third, it introduces dynamic parameter adaptation for Hoeffding trees, a method to dynamically adapt the parameters of Hoeffding trees to reduce their energy consumption. The results show an average energy reduction of 23% on the VFDT algorithm. / Scalable resource-efficient systems for big data analytics

Close to 1% of the world's electricity is consumed by computer servers. Given that the increased use of electricity raises costs and damages the environment, optimizing the world's computing infrastructure for power consumption is worthwhile. This thesis is one attempt at such an optimization. In particular, I began by building a cluster of 6 Intel Atom based low-power nodes to perform work analogous to data center clusters. Then, I installed a version of Hadoop modified with a novel power management system on the cluster. The power management system uses different algorithms to determine when to turn off idle nodes in the cluster. Using the experimental cluster running a modified Hadoop installation, I performed a series of experiments. These tests assessed various strategies for choosing nodes to suspend across a variety of workloads. The experiments validated that turning off idle nodes can yield power savings. While my experimental procedure caused the apparent throughput to significantly decrease, I argue that using more realistic workloads would have yielded much better throughput with slightly reduced power consumption. Additionally, my analysis of the results, show that the percentage power savings in a larger, more realistically sized cluster would be higher than shown in my experiments.

Problemas de escalonamento cuja função objetivo é o consumo de energia tem sido cada vez mais estudados. Neste trabalho, estudamos o problema conhecido, em inglês, por Dynamic Speed Scaling, um problema de escalonamento de tarefas bem definidas em processadores de velocidade variável, cujo consumo de energia é função da velocidade. Além disso, relacionamos este problema com outro conhecido como MOSP, sigla em inglês para Multi-Organization Scheduling Problem. Neste, queremos escalonar tarefas de múltiplas organizações independentes respeitando certas restrições individuais. Provamos, neste trabalho, que este novo problema é NP-Completo e desenvolvemos várias heurísticas eficientes cujos testes experimentais mostram economia de energia significativa. / We studied, in this work, the problem of scheduling a set of well-defined tasks in a variable speed processor with the objective of minimizing the energy consumption, that is given as a function of the processor\'s speed, field known as Dynamic Speed Scaling. Also, we relate this problem to another known as MOSP (Multi-Organization Scheduling Problem), problem in which several independent organizations share tasks and resources to achieve a better global solution, but also respecting selfish restrictions. For the combined problem, we show that it is NP-Complete and designed several efficient heuristics that achieves good results in a experimental setup.