Master of Science (M.S.), University of Dayton, 2016, Electrical Engineering

Today's consumer electronics have a continuous demand on the improvement of the energy density of battery technology. The goal of this work is to develop technology for a high energy - density storage system by concurrently improving Li-ion technology by reducing the weight and thickness of a Li metal anode and the investigation of a wireless battery management system. The wireless battery management system is a novel approach that eliminates the need for wired communication lines in a battery management system. This system has presented multiple advantages and disadvantages over the traditional battery management system, however, it has proven capable of reproducing results taken by a commercial system.

Committee: Guru Subramanyam Ph.D. (Committee Co-Chair); Jitendra Kumar Ph.D. (Committee Co-Chair); Vamsy Chodavarapu Ph.D. (Committee Member) Subjects: Electrical Engineering; Energy

Doctor of Philosophy, The Ohio State University, 2004, Computer and Information Science

The advent in both storage architectures and networking technologies has facilitated storage services using networked storage systems. Often, as many researchers and developers are acutely aware of, hardware and architecture developments that purport to improve performance lack synergy with the software systems they were intended to enhance. New developments in both storage architectures and networking technologies have a profound impact on the design and implementation of networked storage software. In this dissertation, we explore the effects of these advances on the development of networked storage software systems. In particular, how these advances influence communication and memory management and how to design new communication and memory management schemes to take advantage of these advances are investigated. This dissertation first focuses on communication and memory management in the transport layer of a cluster file system over InfiniBand to make the most out of InfiniBand benefits. This dissertation then presents an integrated communication buffer and cache management. This integrated management not only eliminates redundant memory copying and multiple buffering which are considered as main performance bottlenecks of networked storage systems in the general-purpose operating systems, but also enables networked storage software to take full advantage of RDMA benefits in emerging network technologies. This dissertation also introduces a buffering scheme to achieve efficient exclusive caching in multi-level cache hierarchy which is often formed in networked storage systems, which makes better use of memory resources in different cache levels. The main conclusion of this dissertation is that using innovative methods to manage communication and memory can significantly improve performance and scalability of a networked storage system. To achieve this requires studying and taking advantage of the new features in the emerging networking technologies and storage a (open full item for complete abstract)

Committee: Dhabaleswar Panda (Advisor) Subjects: Computer Science

Master of Science in Electrical Engineering, University of Toledo, 2012, Electrical Engineering

The development of micro-girds which combine several localized systems into a small power network has drawn recent attention. They can operate either as a self-contained energy network or they can be integrated into a centralized power grid. A variety of technologies have been studied to use solar energy systems as a form of micro-grid to enhance the reliability and performance of the system. However, the operation of these systems is not without problems, and intermittency of the energy from the sun is the major one. This thesis proposes a microcontroller-based solar energy management system which combines battery management and storage technology to address this issue. This approach uses an energy system with a solar panel array, a maximum power point tracking (MPPT) unit, a battery management system, and a bidirectional inverter which is connected to the electric utility grid. Off-peak energy management also is embedded in the system to further increase the economic benefits.

Committee: Thomas Stuart PhD (Committee Chair); Richard Molyet PhD (Committee Member); Junghwan Kim PhD (Committee Member) Subjects: Engineering

Doctor of Philosophy (PhD), Ohio University, 2024, Industrial and Systems Engineering (Engineering and Technology)

With the onset of technology-driven solutions, the warehousing and logistics sectors are witnessing transformative advancements, one of which is the Robotic Compact Storage and Retrieval System (RCSRS). This research presents a comprehensive examination of RCSRS through three interrelated chapters. The first chapter provides an exhaustive literature review, presenting existing findings and gaps in different types of automated storage systems that have been studied, comparing their characteristics, similarities, and differences. The second chapter pioneers the study of robot travel time within RCSRS, introducing an innovative Mixed-Integer Non-Linear Programming (MINLP) model optimized using a Genetic Algorithm (GA) approach. This investigation primarily provides insights like the optimal placement of the Input/Output (I/O) point and the significance of digging time as a critical bottleneck, while also setting the stage for future research directions. Lastly, the third chapter studies the performance of three optimization algorithms in the RCSRS context: Genetic Algorithm (GA), Simulated Annealing (SA), and a novel Greedy Heuristic. This study aims to minimize robot bin moves, recognizing its direct impact on time and energy utilization. Remarkable findings such as the Greedy Heuristic's efficiency for moderate-sized order lists and the SA's aptness for larger order lists have been detailed. Together, these chapters offer an expansive view into RCSRS's potential and the strategies to harness it, contributing valuable insights and methodologies for the warehousing and logistics sectors. The research anticipates fostering advanced RCSRS designs, optimizing operations, and guiding future research in this transformative domain.

Committee: Tao Yuan Dr. (Committee Chair); Dale Masel Dr. (Committee Co-Chair); Vardges Melkonian Dr. (Committee Member); Ashley Metcalf Dr. (Committee Member); Aros-Vera Felipe Dr. (Committee Member) Subjects: Engineering; Robotics

Doctor of Philosophy, The Ohio State University, 2021, Mechanical Engineering

The electrification of commercial medium- and heavy-duty (MD/HD) vehicles presents new challenges in the design and control of the hybrid powertrain. Advances in Li-Ion battery technology have increased the capabilities of the battery pack, providing more electric range and increased power characteristics. However, new approaches and tools are necessary to improve the overall system efficiency, where a single cell technology may not meet the vocation demand. This dissertation explores the use of a convex optimization framework for the optimal design, energy, and thermal management of Li-Ion battery packs. Convex optimization methods provide advantages in terms of problem formulation, computation speed, and the guarantee of a globally optimal solution. Leveraging a convex framework, new hybrid energy storage systems (HESS) are investigated, where high-energy and high-power batteries are combined in a single system to capitalize on the benefits of each technology. The implementation of a HESS introduces complexities in the pack design and energy management, which are presented through a Design Space Exploration comparing single chemistry and hybrid chemistry battery packs. In addition, the optimal control of the battery thermal management system (TMS) is developed using convex optimization, where new methodologies are explored to mitigate the propagation of approximation error that arise during the convexification process.

Committee: Marcello Canova (Advisor); Giorgio Rizzoni (Committee Member); Abhishek Gupta (Committee Member) Subjects: Mechanical Engineering

Doctor of Philosophy, The Ohio State University, 2020, Electrical and Computer Engineering

DC distribution networks have found increased applications in electric automobiles, ships, aircrafts, server farms, and EV charging stations. These networks contain load regulating power electronic converters such as dc-dc and dc-ac converters that act as Constant Power Loads (CPLs). When these CPLs interact with the dc system, they can cause destabilizing effects on the grid due to their negative incremental impedance. Preliminary studies have performed stability analysis of dc distribution systems and proposed passive stabilization and source/load converter level controlstrategies to address the instability issue which does not address all the stability issues of multi-terminal dc distribution systems. In this research, a method to dynamically stabilize CPLs at the point of load by making them behave as adaptive Smart Resistors using high bandwidth power converters and energy storage units has been proposed. By utilizing high bandwidth power converters, these Smart Resistors can work with smart sources to realize ultimate intelligent power networks. This research aims to identify the realization of the smart Resistor concept and its utilization in the control of dc distribution systems. The effect of the Smart Resistor on the stability of various configurations of dc distribution networks are studied. The aims of this research study are as follows: The control strategies needed to achieve the Smart Resistor concept. The trajectory control of the system during voltage and current transients are studied. The energy management of energy storage is also proposed. The interaction of a constant power load powered by an ideal voltage source and a case study of a traction drive system is performed and the small and large-signal stability of the system is analyzed. The stability of a non-ideal source power converter feeding a constant power load. A case study of a section of a turbo-electric aircraft is used to showcase how the CPL (open full item for complete abstract)

Committee: Jin Wang (Advisor); Anant K Agarwal (Committee Member); Julia Zhang (Committee Chair) Subjects: Engineering

Master of Science, Miami University, 2019, Computational Science and Engineering

The main advantages of electric vehicles (EVs) are reducing the production of greenhouse gases, reducing society's reliance on fossil-fuel-based energy infrastructure, and the ability to achieve high-efficiency energy conversion. Super-capacitors (SCs) are introduced into EV's energy storage system to supplement the primary power source, which improves the vehicle's performance. A low-voltage, high-current bidirectional DC/DC converter serves as an intermediary between the SCs and the primary power source connected to the motor drive. Current-fed CLLC converters are characterized by: a lower input current ripple then voltage source converters, extended input voltage range, reduced transformer size, and the ability to achieve soft switching across the entire input voltage range. This research evaluates the performance of a CF-CLLC converter designed with paralleled Si MOSFETs and compares the performance of a CF-CLLC converter, CLLC converter, and cascaded CF-CLLC converter. A hardware prototype was built for an input voltage of 24 V to 96 V and an output voltage of 400 V with a rated power of 2 kW.

Committee: Mark Scott PhD (Advisor); Chi-Hao Cheng PhD (Committee Member); Miao Wang PhD (Committee Member) Subjects: Electrical Engineering

Doctor of Philosophy, University of Toledo, 2017, Electrical Engineering

Climate change has become one of the most important global challenges that both developing and developed nations face in the 21st Century. In the transportation sector, electric vehicles (xEV) have emerged as a viable solution to fight climate change. However, the short longevity of the battery system, and their limited range which depends on the battery performance, remains a drawback . In the electric power sector, renewable energy (solar and wind) have emerged as a strong alternative, but these sources are intermittent and cause fluctuations on the electrical power grid. To solve these issues, renewable energy systems are sometimes coupled with a battery energy storage system (BESS). Most of these large BESS consist of lithium ion batteries because they are becoming more cost effective than other types [2]. However, there is an issue with both of these applications: as the battery ages, the performance of the battery pack is limited because large variations develop between the large number of series connected cells. This requires the use of an electronic equalizer (EQU) to balance the cell voltages. Currently there are two types of equalizers: passive and active. The passive EQU monitors the weakest cell (lowest voltage), and removes charge from the other cells by dissipating their energy as heat through shunt resistors until all cells voltages equal the weakest cell. This leads to energy loss (heating), reduced capacity, and poor performance. In spite of this, the vast majority of users prefer to use passive EQUs because they are cheap. The second type is called an active EQU, and it transfers charge from one cell to another to balance the cell voltages. This results in reduced energy loss, increased capacity, and higher performance for the battery pack. Although active EQUs have high performance, they remain expensive, costing about 10x the cost of passives. As a result, very few users use active equalizers [10]. This research proposes a solution that has (open full item for complete abstract)

Committee: Thomas Stuart PhD (Committee Chair); Alam Mansoor PhD (Committee Member); Mohsin Jamali PhD (Committee Member); Richard Molyet PhD (Committee Member); Matthew Franchetti PhD (Committee Member) Subjects: Electrical Engineering

Doctor of Philosophy, Case Western Reserve University, 2017, Chemical Engineering

For grid-scale energy storage applications, iron-based hybrid flow batteries have advantages of safety, sustainability and low-cost. Still, several challenges such as device lifetime and efficiency have limited their development. In this work, a new type of hydrogen-ferric ion recombination reactor based on catalyzed, three-dimensional felt is proposed in order to maintain chemical balance in the electrolytes and hence improve the battery stability and lifetime. Cyclic voltammetry (CV) and electrochem- ical impedance spectroscopy (EIS) were used to identify a diffusion-limited hydrogen oxidation current near 0.3 psig of hydrogen partial pressure and show that the perfor- mance can be improved with increasing hydrogen pressure up to about P H 2 = 10 psig. Also, pressure-based measurements showed that high rates of hydrogen recombina- tion (greater than 20 mA cm -2 based on the geometric area and greater than 100 mA cm -2 based on the cross-sectional area) were possible using a floating, membrane-less reactor design. A flow battery model that incorporates the hydrogen evolution side-reactions and chemical rebalancing was developed using a system of differential and algebraic equations (DAE). A good agreement between simulated and measured pressure profiles was obtained for an all-iron flow battery operating at ±100 mA cm -2 . Effects of separator porosity and thickness were simulated, showing how increased thickness and reduced porosity can cause higher pH in the negative electrolyte and hence reduced hydrogen generation. Lastly, a new hybrid flow battery based on mixed, lightly acidic electrolytes was investigated. By using the anomalous codeposition (ACD) phenomenon, it was possible to electrodeposit nearly pure zinc from mixed ZnCl 2 -FeCl 2 electrolytes. The cell was shown to provide 17 % higher voltaic efficiency and 40 % higher power density compared to an all-iron battery operating under the same conditions. A zinc-iron chloride flow battery (open full item for complete abstract)

Committee: robert savinell (Committee Chair); jesse wainright (Committee Member); rohan akolkar (Committee Member); gary wnek (Committee Member) Subjects: Chemical Engineering; Chemistry; Energy; Engineering

Doctor of Engineering, Cleveland State University, 2017, Washkewicz College of Engineering

High Intensity Laser Power Beaming is a wireless power transmission technology developed at the Industrial Space Systems Laboratory from 2005 through 2010, in collaboration with the Air Force Research Laboratory to enable remote optical `refueling' of airborne electric micro unmanned air vehicles. Continuous tracking of these air vehicles with high intensity lasers while in-flight for tens of minutes to recharge the on-board battery system is not operationally practical; hence the recharge time must be minimized. This dissertation presents the development and system design optimization of a hybrid electrical energy storage system as a solution to this practical limitation. The solution is based on the development of a high energy density integrated system to capture and store pulsed energy. The system makes use of ultracapacitors to capture the energy at rapid charge rates, while lithium-ion batteries provide the long-term energy density, in order to maximize the duration of operations and minimize the mass requirements. A design tool employing a genetic algorithm global optimizer was developed to select the front-end ultracapacitor elements. The simulation model and results demonstrate the feasibility of the solution. The hybrid energy storage system is also optimized at the system-level for maximum end-to-end power transfer efficiency. System response optimization results and corresponding sensitivity analysis results are presented. Lastly, the ultrafast supply/extended energy storage system is generalized for other applications such as high-power commercial, industrial, and aerospace applications.

Committee: Hanz Richter Ph.D. (Committee Chair); Taysir Nayfeh Ph.D. (Committee Member); Lili Dong Ph.D. (Committee Member); Majid Rashidi Ph.D. (Committee Member); Petru Fodor Ph.D. (Committee Member) Subjects: Electrical Engineering

Doctor of Business Administration, Cleveland State University, 2017, Monte Ahuja College of Business

Organizations often begin knowledge management (KM) efforts by building knowledge repositories to store organizational knowledge to ensure that it may be later retrieved to reuse, share with, and transfer to knowledge workers. The use of such storage/retrieval systems (S/RS) are particularly relevant in preserving and restoring internal organizational knowledge; such implementations support reduced costs associated with knowledge reacquisition, recreation, and reinvention, thus increasing the efficiency of knowledge transfer. Additionally, there is an increased interest in newer uses of S/RS to support large-scale knowledge-bases and knowledge sharing communities. Therefore, it is important for organizations to understand the factors that influence success in S/RS, as generally, KM systems (KMS) initiatives have failed to realize promised results. This study focuses on knowledge flow from the knowledge repository to the knowledge consumer to facilitate and enable knowledge transfer (FEKT). Because of the strong relationship between S/RS processes and technologies and IS/IT, DeLone and McLean's (2003) IS success model serves as the foundation for the S/RS success model, which is modified here to include the complexities inherent in an S/RS. This empirical study presents a model of S/RS success in FEKT and identifies, prioritizes, and weights both the constructs that define S/RS success and the critical success factors (CSF) that influence these success constructs. In addition to informing KM practitioners, this research also addresses a research gap in the KM literature in respect to storage/retrieval systems in facilitating knowledge transfer. Moreover, while prior KMS research has generally assumed an independence in factors and constructs when empirically testing KMS success, this study embraces the notion that real-world factors and constructs are interrelated, intertwined, and interdependent; thus, the analytic network process (ANP) is used as an analytic method (open full item for complete abstract)

Committee: Radha Appan Ph.D. (Committee Chair); Oya Tukel Ph.D. (Committee Member); Timothy Arndt Ph.D. (Committee Member); Birsen Karpak Ph.D. (Committee Member) Subjects: Information Systems

Doctor of Philosophy, The Ohio State University, 2017, Electrical and Computer Engineering

The advent of microgrids has shifted the focus from centralized power generation to a more distributed manner, involving a mix of different distributed energy resources (DERs). Reciprocating engine driven synchronous generators (referred as gensets) are a common DER used for distributed generation. One of the key concerns with such power networks is the aspect of frequency regulation under large disturbances, especially in an islanded mode of operation, without the support of the utility grid. This works looks at possible solution methods for mitigating large frequency disturbances in an islanded microgrid. Due to steep load changes, the gensets undergo large frequency swings and can be even vulnerable to stalling. The benefits of smart loads are analyzed in this work to prevent such occurrence by temporarily reducing the transient overload on gensets. Another solution to mitigate large frequency deviation is the integration of energy storage system (ESS), but the effectiveness depends on its operation as a grid-forming or a grid-following unit. Important metrics such as frequency nadir during load changes in the islanded microgrid are computed to show the usefulness of ESS in islanded microgrids. For this purpose, analytical methods using reduced-order models are developed and found to provide accurate estimates of frequency deviations under power system disturbances. Generally, ESS units are interfaced with an inverter and when operated in grid-forming mode can offer desired dynamic frequency behavior in an islanded microgrid. Similarly, other inverter-based DERs can also provide good frequency regulation as they share the larger portion of the transient overload compared to gensets. However, under certain scenarios the inverter-based DERs are found to collapse due to this large transient loading and can bring down the whole microgrid system as a result. A better coordination between the different DERs in a mixed source microgrid is facilitated in this work to gua (open full item for complete abstract)

Committee: Mahesh Illindala (Advisor); Jin Wang (Committee Member); Jiankang Wang (Committee Member); Alexander Lindsey (Committee Member) Subjects: Electrical Engineering

MS, University of Cincinnati, 2014, Engineering and Applied Science: Mechanical Engineering

This thesis describes a novel system that is being developed that utilizes latent thermal energy storage (LTES) to shift residential heating and cooling loads, between 2-4 hr. time periods, away from the electrical power grid (during the utilities' peak demand period) for the main purpose of residential demand-side-management. More and more utilities are now offering residential time-of-day rates (and load interrupt programs) to help improve their load factor as a means to curtail the building of new power generating stations, and will only increase in time with greater implementation and the enabling use of smart meters. TES is ideally suited to capitalize on this fact and stands ready in this proposed new system; running the HVAC equipment during the time of excess system capacity and storing the “hot or cold energy created” in the PCM for later use during the peak system demand period will improve the system's load. This thesis describes the proposed system and the equipment layout along with its operating strategy. In addition to being modular in design and thus allowing for all different size homes, another major key feature of the proposed system is that it is of the “plug-in” type which utilizes the current cooling and heating hardware of the existing home, and as such, is equally applicable to new home construction or retrofits. This thesis also presents the economics of the system and potential benefits to the home owner, more specifically, simple calculations are given showing the estimated monthly operating cost savings when using this TES system with residential time-of-day (TD) rates, over that of the home operating without TES and running on the standard residential service (RS) rate structure. This thesis document provides the detailed mathematical formulation for the solution of planar moving boundary problems using enhanced enthalpy method with given fixed temperature and insulated boundary conditions. The solution methodology and results, obt (open full item for complete abstract)

Committee: Michael Kazmierczak Ph.D. (Committee Chair); Ahmed Elgafy Ph.D. (Committee Member); Frank Gerner Ph.D. (Committee Member) Subjects: Mechanics

Doctor of Philosophy, The Ohio State University, 2013, Microbiology

Salmonella enterica is the leading cause of death from foodborne illness in the United States. The mechanisms of its virulence have been well studied but key details have remained elusive. Salmonella expresses a type 1 fimbriae when transiting through the distal small intestine which enables it to bind to mannosylated glycoproteins present on the surface of intestinal epithelial cells. This allows the bacterium to dock to the cell utilizing the Salmonella pathogenicity island 1 encoded type three secretion system (SPI1-T3SS). Upon docking, Salmonella injects effector proteins into the host cell interrupting host cell function. This leads to an actin rearrangement at the site where the bacterium is docked which causes the host cell to engulf the bacterium in a process termed macropinocytosis. These processes are regulated tightly to ensure optimal expression at the time of invasion. The BarA/SirA two-component system in coordination with the Csr system are responsible for the regulation of these virulence traits. BarA is a sensor kinase that monitors the environment for the appropriate signal to start the virulence cascade. Though the signal remains elusive it is hypothesized to be the short chain fatty acids formate and acetate which are found in the distal ileum. Upon sensing these signals BarA phosphorylates SirA, which in turn can activate iii its target genes. To date only two direct targets have been identified for SirA, the small RNA’s csrB and csrC. These small non-coding RNAs are the antagonists of the RNA binding protein CsrA. Typically CsrA binds its target transcripts and prevents translation by blocking the ribosome. Through our studies we have attempted to unravel the pathways by which SirA and CsrA regulate the SPI1-T3SS and type 1 fimbriae to lead to successful invasion of the host cell. We show here that SirA has no direct binding targets in either SPI1 or the fim operon. SirA regulates SPI1 in an indirect manner through the activation of (open full item for complete abstract)

Committee: Brian Ahmer Ph.D. (Advisor); Larry Schlesinger MD/Ph.D. (Committee Member); John Gunn Ph.D. (Committee Member); Robert Munson Ph.D. (Committee Member) Subjects: Genetics; Microbiology

Master of Science, The Ohio State University, 2011, Mechanical Engineering

Recovering and storing a vehicle's kinetic energy during deceleration and the subsequent use of the stored energy during acceleration has lead to significant increases in vehicle efficiency. Current production hybrid electric vehicles (HEVs) convert the energy and store it using electric machines and electro-chemical batteries. While these systems can be configured to provide substantial benefits in addition to kinetic energy recovery, significant limitations exist which hinder the performance and market penetration. Converting mechanical energy to electricity then storing it chemically leads to considerable losses during storage. The path must be followed in the opposite direction during release, compounding the losses. Current HEV batteries, while very effective at storing large quantities of energy, have longevity driven power limitations which drive up cost and weight. As a result of these limitations, investigations have been made into alternative means to recover and store kinetic energy on board vehicles. This thesis investigates two such methods of energy recovery and storage, a hydraulic system with accumulator energy storage and a purely mechanical system with flywheel energy storage. Both systems are of parallel hybrid architecture and offer high power capacity at relatively low cost. The hydraulic system consists of a pump/motor to convert mechanical work to fluid power and a high-pressure accumulator to store the energy. The mechanical system transmits the vehicle's kinetic energy to a flywheel through changing the ratio of a continuously variable transmission linked between the flywheel and the drivetrain. System dynamics models are created for each of the systems' components and coupled to allow for analysis over simulated drive cycles. An iterative design method is proposed for both the hydraulic and mechanical systems, based on drive cycle analysis, performance in simulation, and system properties, such as mass and estimated cost. The systems are co (open full item for complete abstract)

Committee: Marcello Canova PhD (Advisor); Yann Guezennec PhD (Committee Member) Subjects: Automotive Engineering; Mechanical Engineering

Doctor of Philosophy, The Ohio State University, 2010, Computer Science and Engineering

As an emerging storage technology, Flash Memory based Solid State Drive (SSD) has shown a high potential to fundamentally change the existing Hard Disk Drive (HDD) based storage systems. Unlike conventional magnetic disks, SSD is built on semiconductor chips and has no mechanical components (e.g. rotating disk platters). This architectural difference brings many attractive technical features, such as high random data access performance and low power consumption. Most importantly, these unique features could potentially address the long-existing technical limitations of conventional magnetic disks. Due to this reason, SSD has been called a 'pivotal technology' that may completely revolutionize current computer storage systems. On the other hand, SSD also poses several critical challenges to application and system designers. First, due to divergent internal structures, SSD is fundamentally different from rotating media, although they share the same logical and physical interfaces. To date we still lack an insightful understanding about the performance characteristics of SSDs, both the positive and negative sides, and their implications to application and system designers. In this dissertation, we present a thorough experimental study on the unique features of SSDs. Second, although SSDs have shown a great performance potential, especially for handling small and random data accesses, SSDs are much more expensive than conventional hard disks. Even considering the decreasing price trend, the price gap between SSDs and HDDs will not disappear in the near future and it significantly prevents the wide adoption of SSDs in practice, especially in cost-sensitive commercial systems. In this dissertation we present the design and implementation of a hybrid storage system, called Hystor, which integrates both SSDs and HDDs to provide a cost-efficient solution for commercial applications with minimal change to other system components and applications. Third, a unique merit of SSD (open full item for complete abstract)

Committee: Xiaodong Zhang PhD (Committee Chair); Ten-Hwang Lai PhD (Committee Member); Russell Pitzer PhD (Committee Member); Feng Qin PhD (Committee Member) Subjects: Computer Science

Doctor of Philosophy, The Ohio State University, 2008, Computer Science and Engineering

The Internet age has exponentially increased the volume of digital media that is being shared and distributed. Broadband Internet has made technologies such as high quality streaming video on demand possible. Large scale supercomputers also consume and create huge quantities of data. This media and data must be stored, cataloged and retrieved with high-performance. Researching high-performance storage subsystems to meet the I/O demands of applications in modern scenarios is crucial. Advances in microprocessor technology have given rise to relatively cheap off-the-shelf hardware that may be put together as personal computers as well as servers. The servers may be connected together by networking technology to create farms or clusters of workstations (COW). The evolution of COWs has significantly reduced the cost of ownership of high-performance clusters and has allowed users to build fairly large scale machines based on commodity server hardware. As COWs have evolved, networking technologies like InfiniBand and 10 Gigabit Ethernet have also evolved. These networking technologies not only give lower end-to-end latencies, but also allow for better messaging throughput between the nodes. This allows us to connect the clusters with high-performance interconnects at a relatively lower cost. With the deployment of low-cost, high-performance hardware and networking technology, it is increasingly becoming important to design a storage system that can be shared across all the nodes in the cluster. Traditionally, the different components of the file system have been stringed together using network connections. The protocol generally used over the network is TCP/IP. The TCP/IP protocol stack in general has been shown to have poor performance especially for high-performance networks. In this dissertation, we research the problem of designing high-performance communication subsystems for network attached storage (NAS) systems. Specifically, we delve i (open full item for complete abstract)

Committee: Panda Dhabaleswar PhD (Advisor); Ponnuswammy Sadayappan PhD (Committee Member); Feng Qin PhD (Committee Member) Subjects: Computer Science

Master of Science (MS), Ohio University, 1983, Industrial and Manufacturing Systems Engineering (Engineering)

Design of an automated warehouse teaching system

Committee: B Khoshnevis (Advisor) Subjects: Engineering, Industrial

Master of Science (MS), Ohio University, 2004, Computer Science (Engineering)

This research work presents RDSS, a Resource Area Network (RAN)-based Distributed Storage System. It is designed for operational efficiency. Using the RAN architectural pattern, RDSS works on the block level instead of the file level, which increases system flexibility. RDSS adopts directory model routing, lazy master-slave replication, and content hash verification. It uses a Multiple Token Control (MTC) scheme to provide consistent storage services while reducing response delay. It employs a Node Ranking System (NRS) to optimize node selections in terms of stability, integrity, and workload. SPIN (Simple Process/Protocol Modeling Language (PROMELA) Interpreter) simulation results show that RDSS can significantly suppress data transfer of misbehaving nodes. It yields only a slight deviation in terms of individual load share compared to the random selection scheme. Microbenchmarks show that RDSS retrieves 64KB data within 35ms and updates 64KB data within 40ms. The contribution of RDSS is that it provides reliable storage services while achieving high efficiency.

Committee: Chang Liu (Advisor) Subjects:

Master of Technical and Scientific Communication, Miami University, 2003, Technical and Scientific Communication

In January of 2000, I accepted a full-time position with Dell Inc. (formerly known as Dell Computer Corporation) located in Austin, Texas. This report describes the first 18-months of my tenure at Dell and focuses on a major project I completed during this time. I began this project in January 2001 and completed it in March 2001. The other chapters in this report provide a description of Dell Inc., an overview of my internship and my major and minor writing projects, an analysis of the problem-solving model, and some examples of the technical writing assignments that I developed at Dell.

Committee: Jean Lutz (Advisor) Subjects: Computer Science