Combined Cooling, Heating, and Power (CHP) systems have been widely recognized as a key alternative for electric and thermal energy generation because of their outstanding energy efficiency, reduced environmental emissions, and relative independence from centralized power grids. The systems provide simultaneous onsite or near-site electric and thermal energy generation in a single, integrated package. As CHP becomes increasingly popular worldwide and its total capacity increases rapidly, the research on the topics of CHP performance assessment, design, and operational strategy become increasingly important. Following this trend of research activities to improve energy efficiency, environmental emissions, and operational cost, this dissertation focuses on the following aspects: (a) performance evaluation of a CHP system using a transient simulation model; (b) development of a dynamic simulation model of a power generation unit that can be effectively used in transient simulations of CHP systems; (c) investigation of real-time operation of CHP systems based on optimization with respect to operational cost, primary energy consumption, and carbon dioxide emissions; and (d) development of optimal supervisory feedorward control that can provide realistic real-time operation of CHP systems with electric and thermal energy storages using short-term weather forecasting. The results from a transient simulation of a CHP system show that technical and economical performance can be readily evaluated using the transient model and that the design, component selection, and control of a CHP system can be improved using this model. The results from the case studies using optimal real-time operation strategies demonstrate that CHP systems with an energy dispatch algorithm have the potential to yield savings in operational cost, primary energy consumption, and carbon dioxide emissions with respect to a conventional HVAC system. Finally, the results from the case study using a supervisory feedorward control system illustrate that optimal realistic real-time operation of CHP systems with electric and thermal energy storages can be managed by this optimal control using weather forecasting information.

Exploitation of groundwater and excess nutrient runoff are major issues plaguing agriculture and streams in the Lower Mississippi River Basin, and increased irrigation intensity has yielded a proportionate increase in water use. Quantifying the use and effects of conservation practices like on-farm water storage (OFWS) systems will justify continued adoption of these practices to mitigate groundwater decline and nutrient runoff. Since 2010, over 795 hectares of surface water storage has been added in the Big Sunflower River Watershed, and analysis of aquifer saturation shows a recent rise in the water table and a decrease in seasonal water table drawdowns. Modeling pre- and post- construction conditions of a small agricultural subwatershed shows little difference in runoff at the watershed outlet after the construction of an OFWS system, but field monitoring and modeling show more water retained within the system and the importance of management to maximize the benefits of conservation practices.

Thermal energy storage systems as an integral part of concentrated solar power plants improve the performance of the system by mitigating the mismatch between the energy supply and the energy demand. Using a phase change material (PCM) to store energy increases the energy density, hence, reduces the size and cost of the system. However, the performance is limited by the low thermal conductivity of the PCM, which decreases the heat transfer rate between the heat source and PCM, which therefore prolongs the melting, or solidification process, and results in overheating the interface wall. To address this issue, heat pipes are embedded in the PCM to enhance the heat transfer from the receiver to the PCM, and from the PCM to the heat sink during charging and discharging processes, respectively. In the current study, the thermal-fluid phenomenon inside a heat pipe was investigated. The heat pipe network is specifically configured to be implemented in a thermal energy storage unit for a concentrated solar power system. The configuration allows for simultaneous power generation and energy storage for later use. The network is composed of a main heat pipe and an array of secondary heat pipes. The primary heat pipe has a disk-shaped evaporator and a disk-shaped condenser, which are connected via an adiabatic section. The secondary heat pipes are attached to the condenser of the primary heat pipe and they are surrounded by PCM. The other side of the condenser is connected to a heat engine and serves as its heat acceptor. The applied thermal energy to the disk-shaped evaporator changes the phase of working fluid in the wick structure from liquid to vapor. The vapor pressure drives it through the adiabatic section to the condenser where the vapor condenses and releases its heat to a heat engine. It should be noted that the condensed working fluid is returned to the evaporator by the capillary forces of the wick. The extra heat is then delivered to the phase change material through the secondary heat pipes. During the discharging process, secondary heat pipes serve as evaporators and transfer the stored energy to the heat engine. Due to the different geometry of the heat pipe network, a new numerical procedure was developed. The model is axisymmetric and accounts for the compressible vapor flow in the vapor chamber as well as heat conduction in the wall and wick regions. Because of the large expansion ratio from the adiabatic section to the primary condenser, the vapor flow leaving the adiabatic pipe section of the primary heat pipe to the disk-shaped condenser behaves similarly to a confined jet impingement. Therefore, the condensation is not uniform over the main condenser. The feature that makes the numerical procedure distinguished from other available techniques is its ability to simulate non-uniform condensation of the working fluid in the condenser section. The vapor jet impingement on the condenser surface along with condensation is modeled by attaching a porous layer adjacent to the condenser wall. This porous layer acts as a wall, lets the vapor flow to impinge on it, and spread out radially while it allows mass transfer through it. The heat rejection via the vapor condensation is estimated from the mass flux by energy balance at the vapor-liquid interface. This method of simulating heat pipe is proposed and developed in the current work for the first time. Laboratory cylindrical and complex heat pipes and an experimental test rig were designed and fabricated. The measured data from cylindrical heat pipe were used to evaluate the accuracy of the numerical results. The effects of the operating conditions of the heat pipe, heat input, and portion of heat transferred to the phase change material, main condenser geometry, primary heat pipe adiabatic radius and its location as well as secondary heat pipe configurations have been investigated on heat pipe performance. The results showed that in the case with a tubular adiabatic section in the center, the complex interaction of convective and viscous forces in the main condenser chamber, caused several recirculation zones to form in this region, which made the performance of the heat pipe convoluted. The recirculation zone shapes and locations affected by the geometrical features and the heat input, play an important role in the condenser temperature distributions. The temperature distributions of the primary condenser and secondary heat pipe highly depend on the secondary heat pipe configurations and main condenser spacing, especially for the cases with higher heat inputs and higher percentages of heat transfer to the PCM via secondary heat pipes. It was found that changing the entrance shape of the primary condenser and the secondary heat pipes as well as the location and quantity of the secondary heat pipes does not diminish the recirculation zone effects. It was also concluded that changing the location of the adiabatic section reduces the jetting effect of the vapor flow and curtails the recirculation zones, leading to higher average temperature in the main condenser and secondary heat pipes. The experimental results of the conventional heat pipe are presented, however the data for the heat pipe network is not included in this dissertation. The results obtained from the experimental analyses revealed that for the transient operation, as the heat input to the system increases and the conditions at the condenser remains constant, the heat pipe operating temperature increases until it reaches another steady state condition. In addition, the effects of the working fluid and the inclination angle were studied on the performance of a heat pipe. The results showed that in gravity-assisted orientations, the inclination angle has negligible effect on the performance of the heat pipe. However, for gravity-opposed orientations, as the inclination angle increases, the temperature difference between the evaporator and condensation increases which results in higher thermal resistance. It was also found that if the heat pipe is under-filled with the working fluid, the capillary limit of the heat pipe decreases dramatically. However, overfilling of the heat pipe with working fluid degrades the heat pipe performance due to interfering with the evaporation-condensation mechanism. / Mechanical Engineering

Low-latency, high throughput mechanisms to retrieve data become increasingly crucial as the cyber and cyber-physical systems pour out increasing amounts of data that often must be analyzed in an online manner. Generally, as the data volume increases, the marginal utility of an ``average'' data item tends to decline, which requires greater effort in identifying the most valuable data items and making them available with minimal overhead. We believe that data analytics driven mechanisms have a big role to play in solving this needle-in-the-haystack problem. We rely on the claim that efficient pattern discovery and description, coupled with the observed predictability of complex patterns within many applications offers significant potential to enable many I/O optimizations. Our research covers exploitation of storage hierarchy for data driven caching and tiering, reduction of distance between data and computations, removing redundancy in data, using sparse representations of data, the impact of data access mechanisms on resilience, energy consumption, storage usage, and the enablement of new classes of data driven applications. For caching and prefetching, we offer a powerful model that separates the process of access prediction from the data retrieval mechanism. Predictions are made on a data entity basis and used the notions of ``context'' and its aspects such as ``belief'' to uncover and leverage future data needs. This approach allows truly opportunistic utilization of predictive information. We elaborate on which aspects of the context we are using in areas other than caching and prefetching different situations and why it is appropriate in the specified situation. We present in more details the methods we have developed, BeliefCache for data driven caching and prefetching and AVSC for pattern mining based compression of data. In BeliefCache, using a belief, an aspect of context representing an estimate of the probability that the storage element will be needed, we developed modular framework BeliefCache, to make unified informed decisions about that element or a group. For the workloads we examined we were able to capture complex non-sequential access patterns better than a state-of-the-art framework for optimizing cloud storage gateways. Moreover, our framework is also able to adjust to variations in the workload faster. It also does not require a static workload to be effective since modular framework allows for discovering and adapting to the changes in the workload. In AVSC, using an aspect of context to gauge the similarity of the events, we perform our compression by keeping relevant events intact and approximating other events. We do that in two stages. We first generate a summarization of the data, then approximately match the remaining events with the existing patterns if possible, or add the patterns to the summary otherwise. We show gains over the plain lossless compression for a specified amount of accuracy for purposes of identifying the state of the system and a clear tradeoff in between the compressibility and fidelity. In other mentioned research areas we present challenges and opportunities with the hope that will spur researchers to further examine those issues in the space of rapidly emerging data intensive applications. We also discuss the ideas how our research in other domains could be applied in our attempts to provide high performance data access. / Computer and Information Science

Concerns with emissions, CO2 in particular, and energy resource associated with conventional internal combustion engine (ICE) vehicles is motivating a shift towards more electrified power-trains for road transportation, as well as other transportation applications. The modelling, characterization and design of electrified power-trains, including energy storage technologies, traction machine technologies and their associated power electronics, are discussed in this thesis. Port cranes are a special case of land transportation encompassing many of the power-train objectives found common with road based hybrid electric vehicles; here a port crane system is studied. The power flow for a typical crane loading cycle is analyzed and the value of the energy consumption and saving potential is calculated. Then alternative energy storage applications are considered for hybrid power-train configurations employing diesel engine generators, battery packs, supercapacitors (SCs), and flywheels. A hybrid rubber tyred gantry crane (RTGC) power-train model with power management is developed and the battery-SC hybrid energy storage systems are designed for both short- and long-period operation. The Induction machine (IM) is a popular technology for traction applications. Although many publications discuss IM design to realize a traction torque-speed characteristic, the IM model is studied to determine the main parameters impacting on the machine performance capability at constant torque and extended speed. Based on the model analysis, an IM design procedure for traction applications is proposed which improves machine performance capability. The machine design parameters are normalized in per unit form and hence the proposed design procedure is applicable across different ratings. In the specification and definition of vehicle power-trains, it is common (in industry) to quote data at specific operating conditions, for example, full or fixed battery terminal voltage and system temperature. The interactive influence between energy storage devices and the vehicle system is investigated. Using the all-electric Nissan Leaf power-train as a reference example, the Nissan Leaf traction system is evaluated and performance assessed by considering DC-link voltage variation from battery full state of charge (SoC) to zero SoC and temperature variations typical of an automotive application, showing that the system stated performance is reduced as battery SoC decreases. An alternative traction machine design is proposed to satisfy the vehicle target performance requirements over the complete variation of SoC. The vehicle power-train is then modified with the inclusion of a DC/DC converter between the vehicle battery and DC-link to maintain the traction system DC-link voltage near constant. A supercapacitor system is also considered for improved system voltage management. The trade-offs between the actual Nissan Leaf power-train and the redesigned systems are discussed in terms of electronic and machine packaging, and mitigation of faulted operation at high speeds. Using the Nissan Leaf interior permanent magnet (IPM) machine as the benchmark machine, an example surface permanent magnet (SPM) machine, with same design constraints, is designed and compared with the benchmark IPM machine. The phase voltage distortion of IPM and SPM machines are compared and the mechanisms are revealed. An alternative machine topology with pole shoe rotor is proposed for reduction of machine peak current rating and voltage distortion. The pole shoe topology is common in industrial variable speed drives employing constant torque regimes, but not for traction. Here, the machine with pole shoe rotor is designed to achieve traction performance. The pole shoe concept for vehicle traction is significantly different from existing practice in the electric and hybrid electric automotive industry and thus departure in standard design is a contribution of this thesis. / Thesis / Doctor of Philosophy (PhD)

Due to their tight isolation, low overhead, and efficient packaging of the execution environment, Docker containers have become a prominent solution for deploying modern applications. Consequently, a large amount of Docker images are created and this massive image dataset presents challenges to the registry and container storage infrastructure and so far has remained a largely unexplored area. Hence, there is a need of docker image characterization that can help optimize and improve the storage systems for containerized applications. Moreover, existing deduplication techniques significantly degrade the performance of registries, which will slow down the container startup time. Therefore, there is growing demand for high storage efficiency and high-performance registry storage systems. Last but not least, different storage systems can be integrated with containers as backend storage systems and provide persistent storage for containerized applications. So, it is important to analyze the performance of different backend storage systems and storage drivers and draw out the implications for container storage system design. These above observations and challenges motivate my dissertation. In this dissertation, we aim to improve the flexibility, performance, and efficiency of the storage systems for containerized applications. To this end, we focus on the following three important aspects: Docker images, Docker registry storage system, and Docker container storage drivers with their backend storage systems. Specifically, this dissertation adopts three steps: (1) analyzing the Docker image dataset; (2) deriving the design implications; (3) designing a new storage framework for Docker registries and propose different optimizations for container storage systems. In the first part of this dissertation (Chapter 3), we analyze over 167TB of uncompressed Docker Hub images, characterize them using multiple metrics and evaluate the potential of le level deduplication in Docker Hub. In the second part of this dissertation (Chapter 4), we conduct a comprehensive performance analysis of container storage systems based on the key insights from our image characterizations, and derive several design implications. In the third part of this dissertation (Chapter 5), we propose DupHunter, a new Docker registry architecture, which not only natively deduplicates layers for space savings but also reduces layer restore overhead. DupHunter supports several configurable deduplication modes, which provide different levels of storage efficiency, durability, and performance, to support a range of uses. In the fourth part of this dissertation (Chapter 6), we explore an innovative holistic approach, Chameleon, that employs data redundancy techniques such as replication and erasure-coding, coupled with endurance-aware write offloading, to mitigate wear level imbalance in distributed SSD-based storage systems. This high-performance fash cluster can be used for registries to speedup performance. / Doctor of Philosophy / The amount of Docker images stored in Docker registries is increasing rapidly and present challenges for the underlying storage infrastructures. Before we do any optimizations for the storage system, we should first analyze this big Docker image dataset. To this end, in this dissertation we perform the first large-scale characterization and redundancy analysis of the images and layers stored in the Docker Hub registry. Based on the findings, this dissertation presents a series of practical and efficient techniques, algorithms, optimizations to achieve high performance and flexibility, and space-efficient storage system for containerized applications. The experimental evaluation demonstrates the effectiveness of our optimizations and techniques to make storage systems flexible and space-efficacy.

This dissertation presents the development and application of a naval ship distributed system architecture framework, Architecture Flow Optimization (AFO), Dynamic Architecture Flow Optimization (DAFO), and Energy Storage System (ESS) model in naval ship Concept and Requirements Exploration (CandRE). The particular objective of this dissertation is to determine and assess Energy Storage System (ESS) capacity, charging and discharging capabilities in a complex naval ship system of systems to minimize vulnerability and maximize recoverability and effectiveness. The architecture framework is implemented through integrated Ship Behavior Interaction Models (SBIMs) that include the following: Warfighting Model (WM), Ship Operational Model (OM), Capability Model (CM), and Dynamic Architecture Flow Optimization (DAFO). These models provide a critical interface between logical, physical, and operational architectures, quantifying warfighting and propulsion capabilities through system measures of performance at specific capability nodes. This decomposition greatly simplifies the Mission, Power, and Energy System (MPES) design process for use in CandRE. AFO and DAFO are network-based, linear programming optimization methods used to design and analyze MPESs at a sufficient level of detail to understand system energy flow, define MPES architecture and sizing, model operations, reduce system vulnerability and improve system effectiveness and recoverability with ESS capabilities. AFO incorporates system topologies, energy coefficient component models, preliminary arrangements, and (nominal and damaged) steady state scenarios to minimize the energy flow cost required to satisfy all operational scenario demands and constraints. The refined DAFO applies the same principles as AFO, but adds two more capabilities, Propulsion and ESS charging, and maximizes effectiveness at each scenario timestep. DAFO also integrates with a warfighting model, operational model, and capabilities model that quantify the performance of tasks enabled by capabilities through system measures of performance at specific capability nodes. This dissertation provides a description of the design tools developed to implement these processes and methods, including a ship synthesis model, hullform exploration, MPES explorations and objective attribute metrics for cost, effectiveness and risk, using design of experiments (DOEs) response surface models (RSMs) and Energy Storage System (ESS) applications. / Doctor of Philosophy / This dissertation presents the development and application of a naval ship distributed system architecture framework, Architecture Flow Optimization (AFO), Dynamic Architecture Flow Optimization (DAFO), and Energy Storage System (ESS) design in naval ship Concept and Requirements Exploration (CandRE). The particular objective of this dissertation is to determine and assess Energy Storage System (ESS) capacity, charging and discharging capabilities in a complex naval ship system of systems to minimize vulnerability and maximize recoverability and effectiveness. The architecture framework is implemented through integrated Ship Behavior Interaction Models (SBIMs) that include the following: Warfighting Model (WM), Ship Operational Model (OM), Capability Model (CM), and Dynamic Architecture Flow Optimization (DAFO). These models provide a critical interface between logical, physical, and operational architectures, quantifying warfighting and propulsion capabilities through system measures of performance at specific capability nodes. This decomposition greatly simplifies the Mission, Power, and Energy System (MPES) design process for use in CandRE. AFO and DAFO are network-based, linear programming optimization methods used to design and analyze MPESs at a sufficient level of detail to understand system energy flow, define MPES architecture and sizing, model operations, reduce system vulnerability and improve system effectiveness and recoverability with ESS capabilities. AFO incorporates system topologies, energy coefficient component models, preliminary arrangements, and (nominal and damaged) steady state scenarios to minimize the energy flow cost required to satisfy all operational scenario demands and constraints. DAFO applies the same principles as AFO, but adds two more capabilities, Propulsion and ESS charging, and maximizes effectiveness at each scenario timestep. DAFO also integrates with a warfighting model, operational model, and capabilities model that quantify the performance of tasks enabled by capabilities through system measures of performance at specific capability nodes. This dissertation provides an overview of the design tools developed to implement these process and methods, including a ship synthesis model, hullform exploration, MPES explorations and objective attribute metrics for cost, effectiveness and risk, using design of experiments (DOEs) response surface models (RSMs) and Energy Storage System (ESS) applications.

Cooling demand constitutes a large portion of total electrical demand for office buildings during peak hours. Deteriorating load factors, increased use of more inefficient and polluting peaking units are the aftermaths of growth in peak demand challenging energy system efficiency and grid reliability. Ice storage technology can help shift this peak cooling demand to off-peak periods. Ice storage reduces or even eliminates chiller operation during peak periods. The objective of the research is to analyze the chiller energy consumption of conventional non-storage and ice storage cooling systems for large and medium-sized office buildings in diverse climate zones. The research also quantifies the peak energy savings as a result of ice storage systems. To accomplish the thesis objectives the Demand Response Quick Assessment Tool (DRQAT) has been used to model and simulate large and medium-sized office buildings in diverse climate zones with non-storage and ice storage cooling systems. Demand Response Quick Assessment Tool (DRQAT) has been developed by LBNL's Demand Response Research Center. It is based on the most popular features and capabilities of EnergyPlus and is downloadable from [1]. The construction and weather files in DRQAT have been modified to incorporate construction standards and weather data for the cities representing the diverse climate zones. The ice storage system's operating and control strategies investigated include full storage and partial storage with storage priority and chiller priority. Research findings indicate that chiller energy consumption for non-storage and ice storage systems depends highly on climatic conditions. The climate zones with hot summers as well as small day and night temperature variations show higher chiller energy consumption. The marine climate zone has the lowest chiller energy consumption. The cold/humid climate zone has higher chiller energy consumption than the cold/dry and very cold climate zones. The cold/dry and very cold climate zones have comparable chiller energy consumption. The research findings will help utilities and building owners to quantify the benefits of installing ice storage systems in office buildings located in different climate zones. / Master of Science

The growing disparity in data storage and retrieval needs of modern applications is driving the proliferation of a wide variety of storage systems (e.g., key-value stores, cloud storage services, distributed filesystems, and flash cache, etc.). While extant storage systems are designed and tuned for a specific set of applications targeting a range of workload characteristics, they lack the flexibility in adapting to the ever-changing workload behaviors. Moreover, the complexities in implementing modern storage systems and adapting ever-changing storage requirements present unique opportunities and engineering challenges. In this dissertation, we design and develop a series of novel data management and storage systems solutions by applying a simple yet effective rule---workload awareness. We find that simple workload-aware data management strategies are effective in improving the efficiency of modern storage systems, sometimes by an order of magnitude. The first two works tackle the data management and storage space allocation issues at distributed and cloud storage level, while the third work focuses on low-level data management problems in the local storage system, which many high-level storage/data-intensive applications rely on. In the first part of this dissertation (Chapter 3), we propose and develop MBal, a high-performance in-memory object caching framework with adaptive multi-phase load balancing, which supports not only horizontal (scale-out) but vertical (scale-up) scalability as well. MBal is able to make efficient use of available resources in the cloud through its fine-grained, partitioned, lockless design. In the second part of this dissertation (Chapter 4 and Chapter5), we design and build CAST (Chapter 4), a Cloud Analytics Storage Tiering solution that cloud tenants can use to reduce monetary cost and improve performance of analytics workloads. The approach takes the first step towards providing storage tiering support for data analytics in the cloud. Furthermore, we propose a hybrid cloud object storage system (Chapter 5) that could effectively engage both the cloud service providers and cloud tenants via a novel dynamic pricing mechanism. In the third part of this dissertation (Chapter 6), targeting local storage, we explore offline algorithms for flash caching in terms of both hit ratio and flash lifespan. We design and implement a multi-stage heuristic by synthesizing several techniques that manage data at the granularity of a flash erasure unit (which we call a container) to approximate the offline optimal algorithm. In the fourth part of this dissertation (Chapter 7), we are focused on how to enable fast prototyping of efficient distributed key-value stores targeting a proxy-based layered architecture. In this work, we design and build {con}, a framework that significantly reduce the engineering effort required to build a full-fledged distributed key-value store. Our dissertation shows that simple workload-aware data management strategies can bring huge benefit in terms of both efficiency (i.e., performance, monetary cost, etc.) and flexibility (i.e., ease-of-use, ease-of-deployment, programmability, etc.). The principles of leveraging workload dynamicity and storage heterogeneity can be used to guide next-generation storage system software design, especially when being faced with new storage hardware technologies. / Ph. D.

Reliability modelling of RAID storage systems with its various components such as RAID controllers, enclosures, expanders, interconnects and disks is important from a storage system designer's point of view. A model that can express all the failure characteristics of the whole RAID storage system can be used to evaluate design choices, perform cost reliability trade-offs and conduct sensitivity analyses. We present a reliability model for RAID storage systems where we try to model all the components as accurately as possible. We use several state-space reduction techniques, such as aggregating all in-series components and hierarchical decomposition, to reduce the size of our model. To automate computation of reliability, we use the PRISM model checker as a CTMC solver where appropriate. Initially, we assume a simple 3-state disk reliability model with independent disk failures. Later, we assume a Weibull model for the disks; we also consider a correlated disk failure model to check correspondence with the field data available. For all other components in the system, we assume exponential failure distribution. To use the CTMC solver, we approximate the Weibull distribution for a disk using sum of exponentials and we first confirm that this model gives results that are in reasonably good agreement with those from the sequential Monte Carlo simulation methods for RAID disk subsystems. Next, our model for whole RAID storage systems (that includes, for example, disks, expanders, enclosures) uses Weibull distributions and, where appropriate, correlated failure modes for disks, and exponential distributions with independent failure modes for all other components. Since the CTMC solver cannot handle the size of the resulting models, we solve such models using hierarchical decomposition technique. We are able to model fairly large configurations with upto 600 disks using this model. We can use such reasonably complete models to conduct several "what-if" analyses for many RAID storage systems of interest. Our results show that, depending on the configuration, spanning a RAID group across enclosures may increase or decrease reliability. Another key finding from our model results is that redundancy mechanisms such as multipathing is beneficial only if a single failure of some other component does not cause data inaccessibility of a whole RAID group.