The objective of this research is to characterize and manage lifetime reliability, microarchitectural performance, and power tradeoffs in multicore processors. This dissertation is comprised of three research themes; 1) modeling and simulation method of interacting multicore processor physics, 2) characterization and management of performance and lifetime reliability tradeoff, and 3) extending Amdahl’s Law for understanding lifetime reliability, performance, and energy efficiency of heterogeneous processors. With continued technology scaling, processor operations are increasingly dominated by multiple distinct physical phenomena and their coupled interactions. Understanding these behaviors requires the modeling of complex physical interactions. This dissertation first presents a novel simulation framework that orchestrates interactions between multiple physical models and microarchitecture simulators to enable research explorations at the intersection of application, microarchitecture, energy, power, thermal, and reliability. Using this framework, workload-induced variation of device degradation is characterized, and its impacts on processor lifetime and performance are analyzed. This research introduces a new metric to quantify performance-reliability tradeoff. Lastly, the theoretical models of heterogeneous multicore processors are proposed for understanding performance, energy efficiency, and lifetime reliability consequences. It is shown that these system metrics are governed by Amdahl’s Law and correlated as a function of processor composition, scheduling method, and Amdahl’s scaling factor. This dissertation highlights the importance of multidimensional analysis and extends the scope of microarchitectural studies by incorporating the physical aspects of processor operations and designs.

This study explores ways to determine the energy efficiency of a pyrometallurgical ore reduction plant and measures to improve it. The feasibility of building a commercial plant - that is more energy efficient, has a low energy cost, and can operate independently and cost-effectively of external electricity supply - is determined. The need for energy efficiency is expanded to three questions: how should the energy efficiency of the plant be determined, what is the efficiency of the existing plant and to what level it can be improved. Literature and other relevant sources were consulted. Twenty potential energy conservation measures were identified through a literature study. A multi-criteria decision-making approach resulted in the selection of ten measures for conceptual implementation. The measures ranged from high-efficiency motors, solar power, heat recovery with thermal oil and various heat engines, to pressure recovery with turbo-generators. A case study approach was followed with the energy efficiency of an existing prototype plant the subject being studied. The energy usage of the existing plant and feasible measures to improve the performance were empirically observed. The impact of these measures was modelled and the results of the conceptual implementation determined. Two measures that were implemented during the study are also described and the results reported. The study found that the energy efficiency of the plant could be determined by the ratio of product exergy to input energy. By incorporating a number of energy conservation measures conceptually the internal efficiency of the prototype plant was conceptually improved from the current 17% to 22% and as a result externally supplied electricity reduced by 47%. The results were extrapolated to a future commercial plant and energy efficiencies of 26% on-grid and 21% off-grid predicted. This study suggests that a significant improvement in energy efficiency and energy cost can be achieved by integrating appropriate energy conservation measures into the existing and future plants. / MIng (Mechanical Engineering), North-West University, Potchefstroom Campus, 2014

Concerns about global warming are increasing, hence, the urgency to cut carbon emissions. Reducing energy consumption, including lighting energy, is seen as the primary solution. Yet, solving the environmental factor should not come at the cost of other pillars of sustainable development. Rather, maximizing the total value of the building should be the focus. Maximizing value in the context of lighting entails improving the quality of the lighting. This study has investigated how lighting simulation could help in achieving better lighting quality. The aim of the thesis was defined as to understand the underlying architecture of lighting simulation and obtain an overview of its characteristics and applications as well as to study the use of current simulation tools. The theoretical background of lighting simulation (in the domain of Systems Engineering) was reviewed from the literature. This revealed the missing and imperfect links in the solution-to-value chain. The thesis suggests the use of a new base metric, Retinal Illuminance Map, as a solution, which in combination with black box simulation of a visual system can help repair this incomplete chain. The study of the current lighting simulation tools (Paper 1) revealed that illuminance-based metrics, luminance-based metrics, daylight availability metrics, and glare indexes are the most available performance metrics in existing lighting simulation tools. Based on usability, acceptability, availability, and previous references in the literature six software programs (Radiance, DAYSIM, Evalglare, DIALux, VELUX, and VISSLA) were selected and compared. It was found that no single tool could meet all the needs of a designer, hence, simulation tool(s) should be selected (or combined) according to the requirements of project goals and the stage of design. Building on these studies, applications of lighting simulation were identified and compiled in relation to different aspects, including performance metrics, stages of design, optimization, model integration, BIM, and parametric modeling. To obtain first-hand information about lighting designers’ experience, an online survey was conducted in Sweden (Paper 2). The results showed that lighting simulation programs were widely (90%) used in Sweden for analysis and/or rendering purposes. The majority of lighting designers considered both daylight and artificial light in their design. Factors such as ease-of use, simulation time and training had more weight than accuracy and the diversity of metrics in practitioner’s eyes. Surrogate modeling was identified as a solution for speeding up simulation time, which would also enable exploration of design solution space especially in the early design stage.

<p>U radu je izvr&scaron;ena detaljna analiza uticajnih faktora na energetsku efikasnost prijemnika sunčeve energije u dinamičkim uslovima rada. Identifikovani su parametri koji utiču na povećanje efikasnosti reprezentativnih tipova prijemnika koji se nalaze u komercijalnoj primeni. Promenom geometrijskih parametara konstrukcije i radnih parametara utvrđen je njihov uticaj na energetsku efikasnost a zatim je kreiran model za ocenu povećanja efikasnosti prijemnika sunčeve energije u dinamičkim uslovima rada.</p>

Predation, density and disease affect behavior, morphology and growth. There is a lack of information on how these changes relate to efficiency of energy transfer in anuran larvae, although previous studies suggest that predation should decrease and competition should increase efficiency. Using a 2 x 2 factorial design, we manipulated predation presence and larval density to test how predation risk and density affect energy efficiency. During the experiment, approximately half of the tadpoles were infected by an unknown disease. Neither predation risk nor density affected assimilation or growth efficiency, despite changes in growth and development. Disease, however, decreased gut length and growth efficiency. This study builds on past work on the effects of predation and density on a larval amphibian, but also introduces disease as another factor. Our study suggests that disease may be at least as important if not more important than predation or density in regards to growth efficiency.

As part of an effort to be a world leader in international efforts in reducing atmospheric carbon dioxide levels, the UK Government has set itself ambitious targets to reduce carbon dioxide emissions by 80% relative to 1990 levels by 2050. To meet this target, there is a strong emphasis in reducing carbon emissions from the domestic sector through the reduction of energy consumption in UK households by improving the energy efficiency of the housing stock, and the behaviours of the occupants. The Department of Energy and Climate Change have indicated that Local Authorities in England are potentially to work in partnership with businesses and community organizations to facilitate delivery; and as a promoter of domestic energy efficiency policies. Consultation with 11 Local Authorities across England confirmed that they are lacking a reliable mechanism that can detect areas within their administrative boundaries that are most in need of intervention to improve the energy efficiency of the housing stock. For the year 2008 the regression models demonstrate that geographical variations in the size of the house, median household income, and air temperature account for 64% of the variation in English domestic gas consumption, and that variations in the size of the house, median household income, and proportion of households connected to the national gas grid account for 73% of the variation in domestic electricity consumption. The predicted values from these regression models serve as benchmarks of domestic gas and electricity consumption in England having accounted for household income, house size, house type, tenure, and climatic differences and could be used to identify areas within Local Authorities with higher than expected energy consumption for energy efficiency interventions. These results contribute to the wider academic debate over how best to achieve the overall aims of household CO2 reductions by moving beyond a purely technical or behavioural-based approach to reducing domestic energy consumption.

The overarching goal of this thesis is to investigate network architectures, and find the trade-off between low overall energy use and maintaining the level of quality of service (QoS), or even improve it. The ubiquitous wireless communications environment supports the exploration of different network architectures and techniques, the so-called heterogeneous network. Two kinds of heterogeneous architectures are considered: a combined cellular and femtocell network and a combined cellular, femtocell and Wireless Local Area Network(WLAN) network. This thesis concludes that the investigated heterogeneous networks can significantly reduce the overall power consumption, depending on the uptake of femtocells and WLANs. Also, QoS remains high when the power consumption drops. The main energy saving is from reducing the macrocell base station embodied and operational energy. When QoS is evaluated based on the combined cellular and femtocell architecture, it is suggested that use of resource scheduling for femtocells within the macrocell is crucial since femtocell performance is affected significantly by interference when installed in a co-channel system. Additionally, the femtocell transmission power mode is investigated using either variable power level or a fixed power level. To achieve both energy efficiency and QoS, the choice of system configurations should change according to the density of the femtocell deployment. When combining deployment of femtocells with WLANs, more users are able to experience a higher QoS. Due to increasing of data traffic and smartphone usage in the future, WLANs are more important for offloading data from the macrocell, reducing power consumption and also increasing the bandwidth. The localised heterogeneous network is a promising technique for achieving power efficiency and a high QoS system.

The building sector accounts for 40 % of the energy usage, and to be able to reach the energy reduction goals set within the EU and Sweden, the building sector needs to change toward energy efficiency. The building sector has a lot of energy saving potential, and within the sector, the HVAC- and plumbing section has the greatest saving potential. Since building trade magazines are directed to practitioners within the building sector, and also their main channel for information regarding projects and developments, this study has used qualitative content analysis on articles, regarding energy efficiency, between the years of 2002-2014. Two building magazines and one HVAC- and plumbing magazine were used to cover the field of what issues regarding energy efficiency were mediated to the readers. The study found out that during the first years, the magazines mediated a positive image towards energy efficiency measures, but mentioned little concrete action plans. After the implementation of more stringent laws, the magazines mediated different issues in a more equal spreading, however two different issues battled to be heard. These were concerns regarding the risks of using new methods not tried before, the lack of clear definitions from the authorities and a focus on a holistic perspective that included environmental thinking. The most recent years focused on practical solutions, adopting a holistic perspective that included both buildings and individual behaviors. During the years, the magazines in general framed energy efficiency measures as something positive and mediated the image of that energy efficiency measures would be taken in the future.

De nos jours, de nombreuses entreprises choisissent de plus en plus d'adopter le cloud computing. Plus précisément, en tant que clients, elles externalisent la gestion de leur infrastructure physique vers des centres de données (ou plateformes de cloud computing). La consommation d'énergie est une préoccupation majeure pour la gestion des centres de données (datacenter, DC). Son impact financier représente environ 80% du coût total de possession et l'on estime qu'en 2020, les DCs américains dépenseront à eux seuls environ 13 milliards de dollars en factures énergétiques. Généralement, les serveurs de centres de données sont conçus de manière à atteindre une grande efficacité énergétique pour des utilisations élevées. Pour diminuer le coût de calcul, les serveurs de centre de données devraient maximiser leur utilisation. Afin de lutter contre l'utilisation historiquement faible des serveurs, le cloud computing a adopté la virtualisation des serveurs. Cette dernière permet à un serveur physique d'exécuter plusieurs serveurs virtuels (appelés machines virtuelles) de manière isolée. Avec la virtualisation, le fournisseur de cloud peut regrouper (consolider) l'ensemble des machines virtuelles (VM) sur un ensemble réduit de serveurs physiques et ainsi réduire le nombre de serveurs actifs. Même ainsi, les serveurs de centres de données atteignent rarement des utilisations supérieures à 50%, ce qui signifie qu'ils fonctionnent avec des ensembles de ressources majoritairement inutilisées (appelés «trous»). Ma première contribution est un système de gestion de cloud qui divise ou fusionne dynamiquement les machines virtuelles de sorte à ce qu'elles puissent mieux remplir les trous. Cette solution n'est efficace que pour des applications élastiques, c'est-à-dire des applications qui peuvent être exécutées et reconfigurées sur un nombre arbitraire de machines virtuelles. Cependant, la fragmentation des ressources provient d'un problème plus fondamental. On observe que les applications cloud demandent de plus en plus de mémoire, tandis que les serveurs physiques fournissent plus de CPU. Dans les DC actuels, les deux ressources sont fortement couplées puisqu'elles sont liées à un serveur physique. Ma deuxième contribution est un moyen pratique de découpler la paire CPU-mémoire, qui peut être simplement appliquée à n'importe quel serveur. Ainsi, les deux ressources peuvent varier indépendamment, en fonction de leur demande. Ma troisième et ma quatrième contribution montrent un système pratique qui exploite la deuxième contribution. La sous-utilisation observée sur les serveurs physiques existe également pour les machines virtuelles. Il a été démontré que les machines virtuelles ne consomment qu'une petite fraction des ressources allouées car les clients du cloud ne sont pas en mesure d'estimer correctement la quantité de ressources nécessaire à leurs applications. Ma troisième contribution est un système qui estime la consommation de mémoire (c'est-à-dire la taille du working set) d'une machine virtuelle, avec un surcoût faible et une grande précision. Ainsi, nous pouvons maintenant consolider les machines virtuelles en fonction de la taille de leur working set (plutôt que leur mémoire réservée). Cependant, l'inconvénient de cette approche est le risque de famine de mémoire. Si une ou plusieurs machines virtuelles ont une forte augmentation de la demande en mémoire, le serveur physique peut manquer de mémoire. Cette situation n'est pas souhaitable, car la plate-forme cloud est incapable de fournir au client la mémoire qu'il a payée. Finalement, ma quatrième contribution est un système qui permet à une machine virtuelle d'utiliser la mémoire distante fournie par un autre serveur du rack. Ainsi, dans le cas d'un pic de la demande en mémoire, mon système permet à la VM d'allouer de la mémoire sur un serveur physique distant. / Nowadays, many organizations choose to increasingly implement the cloud computing approach. More specifically, as customers, these organizations are outsourcing the management of their physical infrastructure to data centers (or cloud computing platforms). Energy consumption is a primary concern for datacenter (DC) management. Its cost represents about 80% of the total cost of ownership and it is estimated that in 2020, the US DCs alone will spend about $13 billion on energy bills. Generally, the datacenter servers are manufactured in such a way that they achieve high energy efficiency at high utilizations. Thereby for a low cost per computation all datacenter servers should push the utilization as high as possible. In order to fight the historically low utilization, cloud computing adopted server virtualization. The latter allows a physical server to execute multiple virtual servers (called virtual machines) in an isolated way. With virtualization, the cloud provider can pack (consolidate) the entire set of virtual machines (VMs) on a small set of physical servers and thereby, reduce the number of active servers. Even so, the datacenter servers rarely reach utilizations higher than 50% which means that they operate with sets of longterm unused resources (called 'holes'). My first contribution is a cloud management system that dynamically splits/fusions VMs such that they can better fill the holes. This solution is effective only for elastic applications, i.e. applications that can be executed and reconfigured over an arbitrary number of VMs. However the datacenter resource fragmentation stems from a more fundamental problem. Over time, cloud applications demand more and more memory but the physical servers provide more an more CPU. In nowadays datacenters, the two resources are strongly coupled since they are bounded to a physical sever. My second contribution is a practical way to decouple the CPU-memory tuple that can simply be applied to a commodity server. Thereby, the two resources can vary independently, depending on their demand. My third and my forth contribution show a practical system which exploit the second contribution. The underutilization observed on physical servers is also true for virtual machines. It has been shown that VMs consume only a small fraction of the allocated resources because the cloud customers are not able to correctly estimate the resource amount necessary for their applications. My third contribution is a system that estimates the memory consumption (i.e. the working set size) of a VM, with low overhead and high accuracy. Thereby, we can now consolidate the VMs based on their working set size (not the booked memory). However, the drawback of this approach is the risk of memory starvation. If one or multiple VMs have an sharp increase in memory demand, the physical server may run out of memory. This event is undesirable because the cloud platform is unable to provide the client with the booked memory. My fourth contribution is a system that allows a VM to use remote memory provided by a different rack server. Thereby, in the case of a peak memory demand, my system allows the VM to allocate memory on a remote physical server.

This thesis aims to examine energy demand within UK industry and assess the improvement potential available through efficiency measures. The techniques employed throughout the work have been mainly engineering based, drawing on thermodynamics. Alongside this approach, an assessment of drivers and barriers to the technical potential was undertaken. Data availability was a key challenge in the current work. The variety in energy uses meant the use of publically available datasets was limited. A database was constructed utilising site level emissions data, and employed a subsector disaggregation that facilitated energy analysis. The database was used for an analysis of waste heat recovery options. Opportunities were identified in low temperature recovery, heat-ta-power technology, and the transport of heat. Each of these options would require further research and support to be fully realised. It was found that splitting the industrial sector into an energy-intensive and non-energy- intensive subsector, where the grouping was based on the drivers to energy efficiency, allowed generalisations to be made regarding future improvement potential. Based on analysis of past trends, it was found that the energy-intensive subsector has limited potential for further efficiency gains through currently used processes. To make significant improvements radical changes in current processes will be required. A study of the energy-intensive Cement subsector concurred with these findings. Future efficiency improvements in this subsector are likely limited without a shift to alternative cement production. The non-energy-intensive subsector was thought to have relatively greater improvement potential through existing processes. The analysis of these processes is limited by lack of data however. An analysis of the non-energy-intensive Food and drink subsector therefore focussed on improvements in supplying low temperature heat, rather than the efficiency of specific processes. Opportunities through improving steam systems, increasing combined heat-and-power use, and the adoption of heat pumps were found to offer similar improvement potentials.