Energy and power aware scheduling algorithm modeling for battery powered embedded systems

Energy management has always been recognized as a challenge in mobile systems, especially in modern OS-based mobile systems where multi-functioning are widely supported. Nowadays, it is common for a mobile system user to run multiple applications simultaneously while having a target battery lifetime in mind for a specific application. Traditional OS-level power management (PM) policies make their best effort to save energy under performance constraint, but fail to guarantee a target lifetime, leaving the painful trading off between the total performance of applications and the target lifetime to the user itself. This thesis provides a new way to deal with the problem. It is advocated that a strong energy-aware PM scheme should first guarantee a user-specified battery lifetime to a target application by restricting the average power of those less important applications, and in addition to that, maximize the total performance of applications without harming the lifetime guarantee. As a support, energy, instead of CPU or transmission bandwidth, should be globally managed as the first-class resource by the OS. As the first-stage work of a complete PM scheme, this thesis presents the energy-based fair queuing scheduling, a novel class of energy-aware scheduling algorithms which, in combination with a mechanism of battery discharge rate restricting, systematically manage energy as the first-class resource with the objective of guaranteeing a user-specified battery lifetime for a target application in OS-based mobile systems. Energy-based fair queuing is a cross-application of the traditional fair queuing in the energy management domain. It assigns a power share to each task, and manages energy by proportionally serving energy to tasks according to their assigned power shares. The proportional energy use establishes proportional share of the system power among tasks, which guarantees a minimum power for each task and thus, avoids energy starvation on any task. Energy-based fair queuing treats all tasks equally as one type and supports periodical time-sensitive tasks by allocating each of them a share of system power that is adequate to meet the highest energy demand in all periods. However, an overly conservative power share is usually required to guarantee the meeting of all time constraints. To provide more effective and flexible support for various types of time-sensitive tasks in general purpose operating systems, an extra real-time friendly mechanism is introduced to combine priority-based scheduling into the energy-based fair queuing. Since a method is available to control the maximum time one time-sensitive task can run with priority, the power control and time-constraint meeting can be flexibly traded off. A SystemC-based test-bench is designed to assess the algorithms. Simulation results show the success of the energy-based fair queuing in achieving proportional energy use, time-constraint meeting, and a proper trading off between them. La gestión de energía en los sistema móviles está considerada hoy en día como un reto fundamental, notándose, especialmente, en aquellos terminales que utilizando un sistema operativo implementan múltiples funciones. Es común en los sistemas móviles actuales ejecutar simultaneamente diferentes aplicaciones y tener, para una de ellas, un objetivo de tiempo de uso de la batería. Tradicionalmente, las políticas de gestión de consumo de potencia de los sistemas operativos hacen lo que está en sus manos para ahorrar energía y satisfacer sus requisitos de prestaciones, pero no son capaces de proporcionar un objetivo de tiempo de utilización del sistema, dejando al usuario la difícil tarea de buscar un compromiso entre prestaciones y tiempo de utilización del sistema. Esta tesis, como contribución, proporciona una nueva manera de afrontar el problema. En ella se establece que un esquema de gestión de consumo de energía debería, en primer lugar, garantizar, para una aplicación dada, un tiempo mínimo de utilización de la batería que estuviera especificado por el usuario, restringiendo la potencia media consumida por las aplicaciones que se puedan considerar menos importantes y, en segundo lugar, maximizar las prestaciones globales sin comprometer la garantía de utilización de la batería. Como soporte de lo anterior, la energía, en lugar del tiempo de CPU o el ancho de banda, debería gestionarse globalmente por el sistema operativo como recurso de primera clase. Como primera fase en el desarrollo completo de un esquema de gestión de consumo, esta tesis presenta un algoritmo de planificación de encolado equitativo (fair queueing) basado en el consumo de energía, es decir, una nueva clase de algoritmos de planificación que, en combinación con mecanismos que restrinjan la tasa de descarga de una batería, gestionen de forma sistemática la energía como recurso de primera clase, con el objetivo de garantizar, para una aplicación dada, un tiempo de uso de la batería, definido por el usuario, en sistemas móviles empotrados. El encolado equitativo de energía es una extensión al dominio de la energía del encolado equitativo tradicional. Esta clase de algoritmos asigna una reserva de potencia a cada tarea y gestiona la energía sirviéndola de manera proporcional a su reserva. Este uso proporcional de la energía garantiza que cada tarea reciba una porción de potencia y evita que haya tareas que se vean privadas de recibir energía por otras con un comportamiento más ambicioso. Esta clase de algoritmos trata a todas las tareas por igual y puede planificar tareas periódicas en tiempo real asignando a cada una de ellas una reserva de potencia que es adecuada para proporcionar la mayor de las cantidades de energía demandadas por período. Sin embargo, es posible demostrar que sólo se consigue cumplir con los requisitos impuestos por todos los plazos temporales con reservas de potencia extremadamente conservadoras. En esta tesis, para proporcionar un soporte más flexible y eficiente para diferentes tipos de tareas de tiempo real junto con el resto de tareas, se combina un mecanismo de planificación basado en prioridades con el encolado equitativo basado en energía. En esta clase de algoritmos, gracias al método introducido, que controla el tiempo que se ejecuta con prioridad una tarea de tiempo real, se puede establecer un compromiso entre el cumplimiento de los requisitos de tiempo real y el consumo de potencia. Para evaluar los algoritmos, se ha diseñado en SystemC un banco de pruebas. Los resultados muestran que el algoritmo de encolado equitativo basado en el consumo de energía consigue el balance entre el uso proporcional a la energía reservada y el cumplimiento de los requisitos de tiempo real.

GreenDisc: a HW/SW energy optimization framework in globally distributed computation

In recent future, wireless sensor networks (WSNs) will experience a broad high-scale deployment (millions of nodes in the national area) with multiple information sources per node, and with very specific requirements for signal processing. In parallel, the broad range deployment of WSNs facilitates the definition and execution of ambitious studies, with a large input data set and high computational complexity. These computation resources, very often heterogeneous and driven on-demand, can only be satisfied by high-performance Data Centers (DCs). The high economical and environmental impact of the energy consumption in DCs requires aggressive energy optimization policies. These policies have been already detected but not successfully proposed. In this context, this paper shows the following on-going research lines and obtained results. In the field of WSNs: energy optimization in the processing nodes from different abstraction levels, including reconfigurable application specific architectures, efficient customization of the memory hierarchy, energy-aware management of the wireless interface, and design automation for signal processing applications. In the field of DCs: energy-optimal workload assignment policies in heterogeneous DCs, resource management policies with energy consciousness, and efficient cooling mechanisms that will cooperate in the minimization of the electricity bill of the DCs that process the data provided by the WSNs.

GreenDisc: a HW/SW energy optimization framework in globally distributed computation

In recent future, wireless sensor networks ({WSNs}) will experience a broad high-scale deployment (millions of nodes in the national area) with multiple information sources per node, and with very specific requirements for signal processing. In parallel, the broad range deployment of {WSNs} facilitates the definition and execution of ambitious studies, with a large input data set and high computational complexity. These computation resources, very often heterogeneous and driven on-demand, can only be satisfied by high-performance Data Centers ({DCs}). The high economical and environmental impact of the energy consumption in {DCs} requires aggressive energy optimization policies. These policies have been already detected but not successfully proposed. In this context, this paper shows the following on-going research lines and obtained results. In the field of {WSNs}: energy optimization in the processing nodes from different abstraction levels, including reconfigurable application specific architectures, efficient customization of the memory hierarchy, energy-aware management of the wireless interface, and design automation for signal processing applications. In the field of {DCs}: energy-optimal workload assignment policies in heterogeneous {DCs}, resource management policies with energy consciousness, and efficient cooling mechanisms that will cooperate in the minimization of the electricity bill of the DCs that process the data provided by the WSNs.

Energy Management and Optimization for Video Decoders based on Functional-oriented Reconfiguration

In recent years, the increasing sophistication of embedded multimedia systems and wireless communication technologies has promoted a widespread utilization of video streaming applications. It has been reported in 2013 that youngsters, aged between 13 and 24, spend around 16.7 hours a week watching online video through social media, business websites, and video streaming sites. Video applications have already been blended into people daily life. Traditionally, video streaming research has focused on performance improvement, namely throughput increase and response time reduction. However, most mobile devices are battery-powered, a technology that grows at a much slower pace than either multimedia or hardware developments. Since battery developments cannot satisfy expanding power demand of mobile devices, research interests on video applications technology has attracted more attention to achieve energy-efficient designs. How to efficiently use the limited battery energy budget becomes a major research challenge. In addition, next generation video standards impel to diversification and personalization. Therefore, it is desirable to have mechanisms to implement energy optimizations with greater flexibility and scalability. In this context, the main goal of this dissertation is to find an energy management and optimization mechanism to reduce the energy consumption of video decoders based on the idea of functional-oriented reconfiguration. System battery life is prolonged as the result of a trade-off between energy consumption and video quality. Functional-oriented reconfiguration takes advantage of the similarities among standards to build video decoders reconnecting existing functional units. If a feedback channel from the decoder to the encoder is available, the former can signal the latter changes in either the encoding parameters or the encoding algorithms for energy-saving adaption. The proposed energy optimization and management mechanism is carried out at the decoder end. This mechanism consists of an energy-aware manager, implemented as an additional block of the reconfiguration engine, an energy estimator, integrated into the decoder, and, if available, a feedback channel connected to the encoder end. The energy-aware manager checks the battery level, selects the new decoder description and signals to build a new decoder to the reconfiguration engine. It is worth noting that the analysis of the energy consumption is fundamental for the success of the energy management and optimization mechanism. In this thesis, an energy estimation method driven by platform event monitoring is proposed. In addition, an event filter is suggested to automate the selection of the most appropriate events that affect the energy consumption. At last, a detailed study on the influence of the training data on the model accuracy is presented. The modeling methodology of the energy estimator has been evaluated on different underlying platforms, single-core and multi-core, with different characteristics of workload. All the results show a good accuracy and low on-line computation overhead. The required modifications on the reconfiguration engine to implement the energy-aware manager have been assessed under different scenarios. The results indicate a possibility to lengthen the battery lifetime of the system in two different use-cases.

Utilizing the Internet of Things to promote energy awareness and efficiency at discrete production processes: Practices and methodology

El actual contexto de fabricación, con incrementos en los precios de la energía, una creciente preocupación medioambiental y cambios continuos en los comportamientos de los consumidores, fomenta que los responsables prioricen la fabricación respetuosa con el medioambiente. El paradigma del Internet de las Cosas (IoT) promete incrementar la visibilidad y la atención prestada al consumo de energía gracias tanto a sensores como a medidores inteligentes en los niveles de máquina y de línea de producción. En consecuencia es posible y sencillo obtener datos de consumo de energía en tiempo real proveniente de los procesos de fabricación, pero además es posible analizarlos para incrementar su importancia en la toma de decisiones. Esta tesis pretende investigar cómo utilizar la adopción del Internet de las Cosas en el nivel de planta de producción, en procesos discretos, para incrementar la capacidad de uso de la información proveniente tanto de la energía como de la eficiencia energética. Para alcanzar este objetivo general, la investigación se ha dividido en cuatro sub-objetivos y la misma se ha desarrollado a lo largo de cuatro fases principales (en adelante estudios). El primer estudio de esta tesis, que se apoya sobre una revisión bibliográfica comprehensiva y sobre las aportaciones de expertos, define prácticas de gestión de la producción que son energéticamente eficientes y que se apoyan de un modo preeminente en la tecnología IoT. Este primer estudio también detalla los beneficios esperables al adoptar estas prácticas de gestión. Además, propugna un marco de referencia para permitir la integración de los datos que sobre el consumo energético se obtienen en el marco de las plataformas y sistemas de información de la compañía. Esto se lleva a cabo con el objetivo último de remarcar cómo estos datos pueden ser utilizados para apalancar decisiones en los niveles de procesos tanto tácticos como operativos. Segundo, considerando los precios de la energía como variables en el mercado intradiario y la disponibilidad de información detallada sobre el estado de las máquinas desde el punto de vista de consumo energético, el segundo estudio propone un modelo matemático para minimizar los costes del consumo de energía para la programación de asignaciones de una única máquina que deba atender a varios procesos de producción. Este modelo permite la toma de decisiones en el nivel de máquina para determinar los instantes de lanzamiento de cada trabajo de producción, los tiempos muertos, cuándo la máquina debe ser puesta en un estado de apagada, el momento adecuado para rearrancar, y para pararse, etc. Así, este modelo habilita al responsable de producción de implementar el esquema de producción menos costoso para cada turno de producción. En el tercer estudio esta investigación proporciona una metodología para ayudar a los responsables a implementar IoT en el nivel de los sistemas productivos. Se incluye un análisis del estado en que se encuentran los sistemas de gestión de energía y de producción en la factoría, así como también se proporcionan recomendaciones sobre procedimientos para implementar IoT para capturar y analizar los datos de consumo. Esta metodología ha sido validada en un estudio piloto, donde algunos indicadores clave de rendimiento (KPIs) han sido empleados para determinar la eficiencia energética. En el cuarto estudio el objetivo es introducir una vía para obtener visibilidad y relevancia a diferentes niveles de la energía consumida en los procesos de producción. El método propuesto permite que las factorías con procesos de producción discretos puedan determinar la energía consumida, el CO2 emitido o el coste de la energía consumida ya sea en cualquiera de los niveles: operación, producto o la orden de fabricación completa, siempre considerando las diferentes fuentes de energía y las fluctuaciones en los precios de la misma. Los resultados muestran que decisiones y prácticas de gestión para conseguir sistemas de producción energéticamente eficientes son posibles en virtud del Internet de las Cosas. También, con los resultados de esta tesis los responsables de la gestión energética en las compañías pueden plantearse una aproximación a la utilización del IoT desde un punto de vista de la obtención de beneficios, abordando aquellas prácticas de gestión energética que se encuentran más próximas al nivel de madurez de la factoría, a sus objetivos, al tipo de producción que desarrolla, etc. Así mismo esta tesis muestra que es posible obtener reducciones significativas de coste simplemente evitando los períodos de pico diario en el precio de la misma. Además la tesis permite identificar cómo el nivel de monitorización del consumo energético (es decir al nivel de máquina), el intervalo temporal, y el nivel del análisis de los datos son factores determinantes a la hora de localizar oportunidades para mejorar la eficiencia energética. Adicionalmente, la integración de datos de consumo energético en tiempo real con datos de producción (cuando existen altos niveles de estandarización en los procesos productivos y sus datos) es esencial para permitir que las factorías detallen la energía efectivamente consumida, su coste y CO2 emitido durante la producción de un producto o componente. Esto permite obtener una valiosa información a los gestores en el nivel decisor de la factoría así como a los consumidores y reguladores. ABSTRACT In today‘s manufacturing scenario, rising energy prices, increasing ecological awareness, and changing consumer behaviors are driving decision makers to prioritize green manufacturing. The Internet of Things (IoT) paradigm promises to increase the visibility and awareness of energy consumption, thanks to smart sensors and smart meters at the machine and production line level. Consequently, real-time energy consumption data from the manufacturing processes can be easily collected and then analyzed, to improve energy-aware decision-making. This thesis aims to investigate how to utilize the adoption of the Internet of Things at shop floor level to increase energy–awareness and the energy efficiency of discrete production processes. In order to achieve the main research goal, the research is divided into four sub-objectives, and is accomplished during four main phases (i.e., studies). In the first study, by relying on a comprehensive literature review and on experts‘ insights, the thesis defines energy-efficient production management practices that are enhanced and enabled by IoT technology. The first study also explains the benefits that can be obtained by adopting such management practices. Furthermore, it presents a framework to support the integration of gathered energy data into a company‘s information technology tools and platforms, which is done with the ultimate goal of highlighting how operational and tactical decision-making processes could leverage such data in order to improve energy efficiency. Considering the variable energy prices in one day, along with the availability of detailed machine status energy data, the second study proposes a mathematical model to minimize energy consumption costs for single machine production scheduling during production processes. This model works by making decisions at the machine level to determine the launch times for job processing, idle time, when the machine must be shut down, ―turning on‖ time, and ―turning off‖ time. This model enables the operations manager to implement the least expensive production schedule during a production shift. In the third study, the research provides a methodology to help managers implement the IoT at the production system level; it includes an analysis of current energy management and production systems at the factory, and recommends procedures for implementing the IoT to collect and analyze energy data. The methodology has been validated by a pilot study, where energy KPIs have been used to evaluate energy efficiency. In the fourth study, the goal is to introduce a way to achieve multi-level awareness of the energy consumed during production processes. The proposed method enables discrete factories to specify energy consumption, CO2 emissions, and the cost of the energy consumed at operation, production and order levels, while considering energy sources and fluctuations in energy prices. The results show that energy-efficient production management practices and decisions can be enhanced and enabled by the IoT. With the outcomes of the thesis, energy managers can approach the IoT adoption in a benefit-driven way, by addressing energy management practices that are close to the maturity level of the factory, target, production type, etc. The thesis also shows that significant reductions in energy costs can be achieved by avoiding high-energy price periods in a day. Furthermore, the thesis determines the level of monitoring energy consumption (i.e., machine level), the interval time, and the level of energy data analysis, which are all important factors involved in finding opportunities to improve energy efficiency. Eventually, integrating real-time energy data with production data (when there are high levels of production process standardization data) is essential to enable factories to specify the amount and cost of energy consumed, as well as the CO2 emitted while producing a product, providing valuable information to decision makers at the factory level as well as to consumers and regulators.

Proactive and reactive thermal aware optimization techniques to minimize the environmental impact of data centers

Los Centros de Datos se encuentran actualmente en cualquier sector de la economía mundial. Están compuestos por miles de servidores, dando servicio a los usuarios de forma global, las 24 horas del día y los 365 días del año. Durante los últimos años, las aplicaciones del ámbito de la e-Ciencia, como la e-Salud o las Ciudades Inteligentes han experimentado un desarrollo muy significativo. La necesidad de manejar de forma eficiente las necesidades de cómputo de aplicaciones de nueva generación, junto con la creciente demanda de recursos en aplicaciones tradicionales, han facilitado el rápido crecimiento y la proliferación de los Centros de Datos. El principal inconveniente de este aumento de capacidad ha sido el rápido y dramático incremento del consumo energético de estas infraestructuras. En 2010, la factura eléctrica de los Centros de Datos representaba el 1.3% del consumo eléctrico mundial. Sólo en el año 2012, el consumo de potencia de los Centros de Datos creció un 63%, alcanzando los 38GW. En 2013 se estimó un crecimiento de otro 17%, hasta llegar a los 43GW. Además, los Centros de Datos son responsables de más del 2% del total de emisiones de dióxido de carbono a la atmósfera. Esta tesis doctoral se enfrenta al problema energético proponiendo técnicas proactivas y reactivas conscientes de la temperatura y de la energía, que contribuyen a tener Centros de Datos más eficientes. Este trabajo desarrolla modelos de energía y utiliza el conocimiento sobre la demanda energética de la carga de trabajo a ejecutar y de los recursos de computación y refrigeración del Centro de Datos para optimizar el consumo. Además, los Centros de Datos son considerados como un elemento crucial dentro del marco de la aplicación ejecutada, optimizando no sólo el consumo del Centro de Datos sino el consumo energético global de la aplicación. Los principales componentes del consumo en los Centros de Datos son la potencia de computación utilizada por los equipos de IT, y la refrigeración necesaria para mantener los servidores dentro de un rango de temperatura de trabajo que asegure su correcto funcionamiento. Debido a la relación cúbica entre la velocidad de los ventiladores y el consumo de los mismos, las soluciones basadas en el sobre-aprovisionamiento de aire frío al servidor generalmente tienen como resultado ineficiencias energéticas. Por otro lado, temperaturas más elevadas en el procesador llevan a un consumo de fugas mayor, debido a la relación exponencial del consumo de fugas con la temperatura. Además, las características de la carga de trabajo y las políticas de asignación de recursos tienen un impacto importante en los balances entre corriente de fugas y consumo de refrigeración. La primera gran contribución de este trabajo es el desarrollo de modelos de potencia y temperatura que permiten describes estos balances entre corriente de fugas y refrigeración; así como la propuesta de estrategias para minimizar el consumo del servidor por medio de la asignación conjunta de refrigeración y carga desde una perspectiva multivariable. Cuando escalamos a nivel del Centro de Datos, observamos un comportamiento similar en términos del balance entre corrientes de fugas y refrigeración. Conforme aumenta la temperatura de la sala, mejora la eficiencia de la refrigeración. Sin embargo, este incremente de la temperatura de sala provoca un aumento en la temperatura de la CPU y, por tanto, también del consumo de fugas. Además, la dinámica de la sala tiene un comportamiento muy desigual, no equilibrado, debido a la asignación de carga y a la heterogeneidad en el equipamiento de IT. La segunda contribución de esta tesis es la propuesta de técnicas de asigación conscientes de la temperatura y heterogeneidad que permiten optimizar conjuntamente la asignación de tareas y refrigeración a los servidores. Estas estrategias necesitan estar respaldadas por modelos flexibles, que puedan trabajar en tiempo real, para describir el sistema desde un nivel de abstracción alto. Dentro del ámbito de las aplicaciones de nueva generación, las decisiones tomadas en el nivel de aplicación pueden tener un impacto dramático en el consumo energético de niveles de abstracción menores, como por ejemplo, en el Centro de Datos. Es importante considerar las relaciones entre todos los agentes computacionales implicados en el problema, de forma que puedan cooperar para conseguir el objetivo común de reducir el coste energético global del sistema. La tercera contribución de esta tesis es el desarrollo de optimizaciones energéticas para la aplicación global por medio de la evaluación de los costes de ejecutar parte del procesado necesario en otros niveles de abstracción, que van desde los nodos hasta el Centro de Datos, por medio de técnicas de balanceo de carga. Como resumen, el trabajo presentado en esta tesis lleva a cabo contribuciones en el modelado y optimización consciente del consumo por fugas y la refrigeración de servidores; el modelado de los Centros de Datos y el desarrollo de políticas de asignación conscientes de la heterogeneidad; y desarrolla mecanismos para la optimización energética de aplicaciones de nueva generación desde varios niveles de abstracción. ABSTRACT Data centers are easily found in every sector of the worldwide economy. They consist of tens of thousands of servers, serving millions of users globally and 24-7. In the last years, e-Science applications such e-Health or Smart Cities have experienced a significant development. The need to deal efficiently with the computational needs of next-generation applications together with the increasing demand for higher resources in traditional applications has facilitated the rapid proliferation and growing of data centers. A drawback to this capacity growth has been the rapid increase of the energy consumption of these facilities. In 2010, data center electricity represented 1.3% of all the electricity use in the world. In year 2012 alone, global data center power demand grew 63% to 38GW. A further rise of 17% to 43GW was estimated in 2013. Moreover, data centers are responsible for more than 2% of total carbon dioxide emissions. This PhD Thesis addresses the energy challenge by proposing proactive and reactive thermal and energy-aware optimization techniques that contribute to place data centers on a more scalable curve. This work develops energy models and uses the knowledge about the energy demand of the workload to be executed and the computational and cooling resources available at data center to optimize energy consumption. Moreover, data centers are considered as a crucial element within their application framework, optimizing not only the energy consumption of the facility, but the global energy consumption of the application. The main contributors to the energy consumption in a data center are the computing power drawn by IT equipment and the cooling power needed to keep the servers within a certain temperature range that ensures safe operation. Because of the cubic relation of fan power with fan speed, solutions based on over-provisioning cold air into the server usually lead to inefficiencies. On the other hand, higher chip temperatures lead to higher leakage power because of the exponential dependence of leakage on temperature. Moreover, workload characteristics as well as allocation policies also have an important impact on the leakage-cooling tradeoffs. The first key contribution of this work is the development of power and temperature models that accurately describe the leakage-cooling tradeoffs at the server level, and the proposal of strategies to minimize server energy via joint cooling and workload management from a multivariate perspective. When scaling to the data center level, a similar behavior in terms of leakage-temperature tradeoffs can be observed. As room temperature raises, the efficiency of data room cooling units improves. However, as we increase room temperature, CPU temperature raises and so does leakage power. Moreover, the thermal dynamics of a data room exhibit unbalanced patterns due to both the workload allocation and the heterogeneity of computing equipment. The second main contribution is the proposal of thermal- and heterogeneity-aware workload management techniques that jointly optimize the allocation of computation and cooling to servers. These strategies need to be backed up by flexible room level models, able to work on runtime, that describe the system from a high level perspective. Within the framework of next-generation applications, decisions taken at this scope can have a dramatical impact on the energy consumption of lower abstraction levels, i.e. the data center facility. It is important to consider the relationships between all the computational agents involved in the problem, so that they can cooperate to achieve the common goal of reducing energy in the overall system. The third main contribution is the energy optimization of the overall application by evaluating the energy costs of performing part of the processing in any of the different abstraction layers, from the node to the data center, via workload management and off-loading techniques. In summary, the work presented in this PhD Thesis, makes contributions on leakage and cooling aware server modeling and optimization, data center thermal modeling and heterogeneityaware data center resource allocation, and develops mechanisms for the energy optimization for next-generation applications from a multi-layer perspective.

Leakage and temperature aware server control for improving energy efficiency in data centers

Reducing the energy consumption for computation and cooling in servers is a major challenge considering the data center energy costs today. To ensure energy-efficient operation of servers in data centers, the relationship among computa- tional power, temperature, leakage, and cooling power needs to be analyzed. By means of an innovative setup that enables monitoring and controlling the computing and cooling power consumption separately on a commercial enterprise server, this paper studies temperature-leakage-energy tradeoffs, obtaining an empirical model for the leakage component. Using this model, we design a controller that continuously seeks and settles at the optimal fan speed to minimize the energy consumption for a given workload. We run a customized dynamic load-synthesis tool to stress the system. Our proposed cooling controller achieves up to 9% energy savings and 30W reduction in peak power in comparison to the default cooling control scheme.

Leakage-aware cooling management for improving server energy efficiency

The computational and cooling power demands of enterprise servers are increasing at an unsustainable rate. Understanding the relationship between computational power, temperature, leakage, and cooling power is crucial to enable energy-efficient operation at the server and data center levels. This paper develops empirical models to estimate the contributions of static and dynamic power consumption in enterprise servers for a wide range of workloads, and analyzes the interactions between temperature, leakage, and cooling power for various workload allocation policies. We propose a cooling management policy that minimizes the server energy consumption by setting the optimum fan speed during runtime. Our experimental results on a presently shipping enterprise server demonstrate that including leakage awareness in workload and cooling management provides additional energy savings without any impact on performance.

Localization and Tracking in Energy-Constrained Networks with Dynamic Connectivity

Systems used for target localization, such as goods, individuals, or animals, commonly rely on operational means to meet the final application demands. However, what would happen if some means were powered up randomly by harvesting systems? And what if those devices not randomly powered had their duty cycles restricted? Under what conditions would such an operation be tolerable in localization services? What if the references provided by nodes in a tracking problem were distorted? Moreover, there is an underlying topic common to the previous questions regarding the transfer of conceptual models to reality in field tests: what challenges are faced upon deploying a localization network that integrates energy harvesting modules? The application scenario of the system studied is a traditional herding environment of semi domesticated reindeer (Rangifer tarandus tarandus) in northern Scandinavia. In these conditions, information on approximate locations of reindeer is as important as environmental preservation. Herders also need cost-effective devices capable of operating unattended in, sometimes, extreme weather conditions. The analyses developed are worthy not only for the specific application environment presented, but also because they may serve as an approach to performance of navigation systems in absence of reasonably accurate references like the ones of the Global Positioning System (GPS). A number of energy-harvesting solutions, like thermal and radio-frequency harvesting, do not commonly provide power beyond one milliwatt. When they do, battery buffers may be needed (as it happens with solar energy) which may raise costs and make systems more dependent on environmental temperatures. In general, given our problem, a harvesting system is needed that be capable of providing energy bursts of, at least, some milliwatts. Many works on localization problems assume that devices have certain capabilities to determine unknown locations based on range-based techniques or fingerprinting which cannot be assumed in the approach considered herein. The system presented is akin to range-free techniques, but goes to the extent of considering very low node densities: most range-free techniques are, therefore, not applicable. Animal localization, in particular, uses to be supported by accurate devices such as GPS collars which deplete batteries in, maximum, a few days. Such short-life solutions are not particularly desirable in the framework considered. In tracking, the challenge may times addressed aims at attaining high precision levels from complex reliable hardware and thorough processing techniques. One of the challenges in this Thesis is the use of equipment with just part of its facilities in permanent operation, which may yield high input noise levels in the form of distorted reference points. The solution presented integrates a kinetic harvesting module in some nodes which are expected to be a majority in the network. These modules are capable of providing power bursts of some milliwatts which suffice to meet node energy demands. The usage of harvesting modules in the aforementioned conditions makes the system less dependent on environmental temperatures as no batteries are used in nodes with harvesters--it may be also an advantage in economic terms. There is a second kind of nodes. They are battery powered (without kinetic energy harvesters), and are, therefore, dependent on temperature and battery replacements. In addition, their operation is constrained by duty cycles in order to extend node lifetime and, consequently, their autonomy. There is, in turn, a third type of nodes (hotspots) which can be static or mobile. They are also battery-powered, and are used to retrieve information from the network so that it is presented to users. The system operational chain starts at the kinetic-powered nodes broadcasting their own identifier. If an identifier is received at a battery-powered node, the latter stores it for its records. Later, as the recording node meets a hotspot, its full record of detections is transferred to the hotspot. Every detection registry comprises, at least, a node identifier and the position read from its GPS module by the battery-operated node previously to detection. The characteristics of the system presented make the aforementioned operation own certain particularities which are also studied. First, identifier transmissions are random as they depend on movements at kinetic modules--reindeer movements in our application. Not every movement suffices since it must overcome a certain energy threshold. Second, identifier transmissions may not be heard unless there is a battery-powered node in the surroundings. Third, battery-powered nodes do not poll continuously their GPS module, hence localization errors rise even more. Let's recall at this point that such behavior is tight to the aforementioned power saving policies to extend node lifetime. Last, some time is elapsed between the instant an identifier random transmission is detected and the moment the user is aware of such a detection: it takes some time to find a hotspot. Tracking is posed as a problem of a single kinetically-powered target and a population of battery-operated nodes with higher densities than before in localization. Since the latter provide their approximate positions as reference locations, the study is again focused on assessing the impact of such distorted references on performance. Unlike in localization, distance-estimation capabilities based on signal parameters are assumed in this problem. Three variants of the Kalman filter family are applied in this context: the regular Kalman filter, the alpha-beta filter, and the unscented Kalman filter. The study enclosed hereafter comprises both field tests and simulations. Field tests were used mainly to assess the challenges related to power supply and operation in extreme conditions as well as to model nodes and some aspects of their operation in the application scenario. These models are the basics of the simulations developed later. The overall system performance is analyzed according to three metrics: number of detections per kinetic node, accuracy, and latency. The links between these metrics and the operational conditions are also discussed and characterized statistically. Subsequently, such statistical characterization is used to forecast performance figures given specific operational parameters. In tracking, also studied via simulations, nonlinear relationships are found between accuracy and duty cycles and cluster sizes of battery-operated nodes. The solution presented may be more complex in terms of network structure than existing solutions based on GPS collars. However, its main gain lies on taking advantage of users' error tolerance to reduce costs and become more environmentally friendly by diminishing the potential amount of batteries that can be lost. Whether it is applicable or not depends ultimately on the conditions and requirements imposed by users' needs and operational environments, which is, as it has been explained, one of the topics of this Thesis.

Proactive and reactive thermal aware optimization techniques to minimize the environmental impact of data centers

Data centers are easily found in every sector of the worldwide economy. They are composed of thousands of servers, serving millions of users globally and 24-7. In the last years, e-Science applications such e-Health or Smart Cities have experienced a significant development. The need to deal efficiently with the computational needs of next-generation applications together with the increasing demand for higher resources in traditional applications has facilitated the rapid proliferation and growing of Data Centers. A drawback to this capacity growth has been the rapid increase of the energy consumption of these facilities. In 2010, data center electricity represented 1.3% of all the electricity use in the world. In year 2012 alone, global data center power demand grep 63% to 38GW. A further rise of 17% to 43GW was estimated in 2013. Moreover, Data Centers are responsible for more than 2% of total carbon dioxide emissions.

Design and Simulation of Deep Nanometer SRAM Cells under Energy, Mismatch, and Radiation Constraints

La fiabilidad está pasando a ser el principal problema de los circuitos integrados según la tecnología desciende por debajo de los 22nm. Pequeñas imperfecciones en la fabricación de los dispositivos dan lugar ahora a importantes diferencias aleatorias en sus características eléctricas, que han de ser tenidas en cuenta durante la fase de diseño. Los nuevos procesos y materiales requeridos para la fabricación de dispositivos de dimensiones tan reducidas están dando lugar a diferentes efectos que resultan finalmente en un incremento del consumo estático, o una mayor vulnerabilidad frente a radiación. Las memorias SRAM son ya la parte más vulnerable de un sistema electrónico, no solo por representar más de la mitad del área de los SoCs y microprocesadores actuales, sino también porque las variaciones de proceso les afectan de forma crítica, donde el fallo de una única célula afecta a la memoria entera. Esta tesis aborda los diferentes retos que presenta el diseño de memorias SRAM en las tecnologías más pequeñas. En un escenario de aumento de la variabilidad, se consideran problemas como el consumo de energía, el diseño teniendo en cuenta efectos de la tecnología a bajo nivel o el endurecimiento frente a radiación. En primer lugar, dado el aumento de la variabilidad de los dispositivos pertenecientes a los nodos tecnológicos más pequeños, así como a la aparición de nuevas fuentes de variabilidad por la inclusión de nuevos dispositivos y la reducción de sus dimensiones, la precisión del modelado de dicha variabilidad es crucial. Se propone en la tesis extender el método de inyectores, que modela la variabilidad a nivel de circuito, abstrayendo sus causas físicas, añadiendo dos nuevas fuentes para modelar la pendiente sub-umbral y el DIBL, de creciente importancia en la tecnología FinFET. Los dos nuevos inyectores propuestos incrementan la exactitud de figuras de mérito a diferentes niveles de abstracción del diseño electrónico: a nivel de transistor, de puerta y de circuito. El error cuadrático medio al simular métricas de estabilidad y prestaciones de células SRAM se reduce un mínimo de 1,5 veces y hasta un máximo de 7,5 a la vez que la estimación de la probabilidad de fallo se mejora en varios ordenes de magnitud. El diseño para bajo consumo es una de las principales aplicaciones actuales dada la creciente importancia de los dispositivos móviles dependientes de baterías. Es igualmente necesario debido a las importantes densidades de potencia en los sistemas actuales, con el fin de reducir su disipación térmica y sus consecuencias en cuanto al envejecimiento. El método tradicional de reducir la tensión de alimentación para reducir el consumo es problemático en el caso de las memorias SRAM dado el creciente impacto de la variabilidad a bajas tensiones. Se propone el diseño de una célula que usa valores negativos en la bit-line para reducir los fallos de escritura según se reduce la tensión de alimentación principal. A pesar de usar una segunda fuente de alimentación para la tensión negativa en la bit-line, el diseño propuesto consigue reducir el consumo hasta en un 20 % comparado con una célula convencional. Una nueva métrica, el hold trip point se ha propuesto para prevenir nuevos tipos de fallo debidos al uso de tensiones negativas, así como un método alternativo para estimar la velocidad de lectura, reduciendo el número de simulaciones necesarias. Según continúa la reducción del tamaño de los dispositivos electrónicos, se incluyen nuevos mecanismos que permiten facilitar el proceso de fabricación, o alcanzar las prestaciones requeridas para cada nueva generación tecnológica. Se puede citar como ejemplo el estrés compresivo o extensivo aplicado a los fins en tecnologías FinFET, que altera la movilidad de los transistores fabricados a partir de dichos fins. Los efectos de estos mecanismos dependen mucho del layout, la posición de unos transistores afecta a los transistores colindantes y pudiendo ser el efecto diferente en diferentes tipos de transistores. Se propone el uso de una célula SRAM complementaria que utiliza dispositivos pMOS en los transistores de paso, así reduciendo la longitud de los fins de los transistores nMOS y alargando los de los pMOS, extendiéndolos a las células vecinas y hasta los límites de la matriz de células. Considerando los efectos del STI y estresores de SiGe, el diseño propuesto mejora los dos tipos de transistores, mejorando las prestaciones de la célula SRAM complementaria en más de un 10% para una misma probabilidad de fallo y un mismo consumo estático, sin que se requiera aumentar el área. Finalmente, la radiación ha sido un problema recurrente en la electrónica para aplicaciones espaciales, pero la reducción de las corrientes y tensiones de los dispositivos actuales los está volviendo vulnerables al ruido generado por radiación, incluso a nivel de suelo. Pese a que tecnologías como SOI o FinFET reducen la cantidad de energía colectada por el circuito durante el impacto de una partícula, las importantes variaciones de proceso en los nodos más pequeños va a afectar su inmunidad frente a la radiación. Se demuestra que los errores inducidos por radiación pueden aumentar hasta en un 40 % en el nodo de 7nm cuando se consideran las variaciones de proceso, comparado con el caso nominal. Este incremento es de una magnitud mayor que la mejora obtenida mediante el diseño de células de memoria específicamente endurecidas frente a radiación, sugiriendo que la reducción de la variabilidad representaría una mayor mejora. ABSTRACT Reliability is becoming the main concern on integrated circuit as the technology goes beyond 22nm. Small imperfections in the device manufacturing result now in important random differences of the devices at electrical level which must be dealt with during the design. New processes and materials, required to allow the fabrication of the extremely short devices, are making new effects appear resulting ultimately on increased static power consumption, or higher vulnerability to radiation SRAMs have become the most vulnerable part of electronic systems, not only they account for more than half of the chip area of nowadays SoCs and microprocessors, but they are critical as soon as different variation sources are regarded, with failures in a single cell making the whole memory fail. This thesis addresses the different challenges that SRAM design has in the smallest technologies. In a common scenario of increasing variability, issues like energy consumption, design aware of the technology and radiation hardening are considered. First, given the increasing magnitude of device variability in the smallest nodes, as well as new sources of variability appearing as a consequence of new devices and shortened lengths, an accurate modeling of the variability is crucial. We propose to extend the injectors method that models variability at circuit level, abstracting its physical sources, to better model sub-threshold slope and drain induced barrier lowering that are gaining importance in FinFET technology. The two new proposed injectors bring an increased accuracy of figures of merit at different abstraction levels of electronic design, at transistor, gate and circuit levels. The mean square error estimating performance and stability metrics of SRAM cells is reduced by at least 1.5 and up to 7.5 while the yield estimation is improved by orders of magnitude. Low power design is a major constraint given the high-growing market of mobile devices that run on battery. It is also relevant because of the increased power densities of nowadays systems, in order to reduce the thermal dissipation and its impact on aging. The traditional approach of reducing the voltage to lower the energy consumption if challenging in the case of SRAMs given the increased impact of process variations at low voltage supplies. We propose a cell design that makes use of negative bit-line write-assist to overcome write failures as the main supply voltage is lowered. Despite using a second power source for the negative bit-line, the design achieves an energy reduction up to 20% compared to a conventional cell. A new metric, the hold trip point has been introduced to deal with new sources of failures to cells using a negative bit-line voltage, as well as an alternative method to estimate cell speed, requiring less simulations. With the continuous reduction of device sizes, new mechanisms need to be included to ease the fabrication process and to meet the performance targets of the successive nodes. As example we can consider the compressive or tensile strains included in FinFET technology, that alter the mobility of the transistors made out of the concerned fins. The effects of these mechanisms are very dependent on the layout, with transistor being affected by their neighbors, and different types of transistors being affected in a different way. We propose to use complementary SRAM cells with pMOS pass-gates in order to reduce the fin length of nMOS devices and achieve long uncut fins for the pMOS devices when the cell is included in its corresponding array. Once Shallow Trench isolation and SiGe stressors are considered the proposed design improves both kinds of transistor, boosting the performance of complementary SRAM cells by more than 10% for a same failure probability and static power consumption, with no area overhead. While radiation has been a traditional concern in space electronics, the small currents and voltages used in the latest nodes are making them more vulnerable to radiation-induced transient noise, even at ground level. Even if SOI or FinFET technologies reduce the amount of energy transferred from the striking particle to the circuit, the important process variation that the smallest nodes will present will affect their radiation hardening capabilities. We demonstrate that process variations can increase the radiation-induced error rate by up to 40% in the 7nm node compared to the nominal case. This increase is higher than the improvement achieved by radiation-hardened cells suggesting that the reduction of process variations would bring a higher improvement.