<p>Modern computing platforms, from servers to mobile devices, demand ever-increasing amounts of memory to keep up with the growing amounts of data they process, and to bridge the widening processor-memory gap. A large and growing fraction of chip area and energy is expended in memories, which face challenges with technology scaling due to increased leakage, process variations, and unreliability. On the other hand, data intensive workloads such as machine learning and data analytics pose increasing demands on memory systems. Consequently, improving the energy-efficiency and performance of memory systems is an important challenge for computing system designers.</p> <p>Spintronic memories, which offer several desirable characteristics - near-zero leakage, high density, non-volatility and high endurance - are of great interest for designing future memory systems. However, these memories are not drop-in replacements for current memory technologies, viz. Static Random Access Memory (SRAM) and Dynamic Random Access Memory (DRAM). They pose unique challenges such as variable access times, and require higher write latency and write energy. This dissertation explores new approaches to improving the energy efficiency of spintronic memory systems.</p> <p>The dissertation first explores the design of approximate memories, in which the need to store and access data precisely is foregone in return for improvements in energy efficiency. This is of particular interest, since many emerging workloads exhibit an inherent ability to tolerate approximations to their underlying computations and data while still producing outputs of acceptable quality. The dissertation proposes that approximate spintronic memories can be realized either by reducing the amount of data that is written to/read from them, or by reducing the energy consumed per access. To reduce memory traffic, the dissertation proposes approximate memory compression, wherein a quality-aware memory controller transparently compresses/decompresses data written to or read from memory. For broader applicability, the quality-aware memory controller can be programmed to specify memory regions that can tolerate approximations, and conforms to a specified error constraint for each such region. To reduce the per-access energy, various mechanisms are identified at the circuit and architecture levels that yield substantial energy benefits at the cost of small probabilities of read, write or retention failures. Based on these mechanisms, a quality-configurable Spin Transfer Torque Magnetic RAM (STT-MRAM) array is designed in which read/write operations can be performed at varying levels of accuracy and energy at runtime, depending on the needs of applications. To illustrate the utility of the proposed quality-configurable memory array, it is evaluated as an L2 cache in the context of a general-purpose processor, and as a scratchpad memory for a domain-specific vector processor.</p> <p>The dissertation also explores the design of caches with Domain Wall Memory (DWM), a more advanced spintronic memory technology that offers unparalleled density arising from a unique tape-like structure. However, this structure also leads to serialized access to the bits in each bit-cell, resulting in increased access latency, thereby degrading overall performance. To mitigate the performance overheads, the dissertation proposes a reconfigurable DWM-based cache architecture that modulates the active bits per tape with minimal overheads depending on the application's memory access characteristics. The proposed cache is evaluated in a general purpose processor and improvements in performance are demonstrated over both CMOS and previously proposed spintronic caches.</p> <p>In summary, the dissertation suggests directions to improve the energy efficiency of spintronic memories and re-affirms their potential for the design of future memory systems.</p>

Made available in DSpace on 2014-06-11T19:22:18Z (GMT). No. of bitstreams: 0 Previous issue date: 2011-05-12Bitstream added on 2014-06-13T18:48:40Z : No. of bitstreams: 1 denadai_d_me_sjrp.pdf: 370065 bytes, checksum: 6db2aacdc27e3eb4ca5f48d683ca1d56 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Neste trabalho provamos a existência de controle aproximado para certos sistemas não-lineares de equaçõ es diferenciais ordinárias de entrada e saída únicas e múltiplas. Utilizamos como técnica funções ou aplicações implícitas globais. / In this work we prove the existence of approximate control for certain nonlinear sys-tems of ordinary differential equations of single-input single-output and multi-input multi-output. We use global implicit functions or mappings.

<p>Unakrsna aproksimativna entropija kvantifikuje međusobnu uređenost<br />dva istomvremeno snimljena vremenska niza. Iako je izvedena iz<br />veoma zastupljene entropije za procenu uređenosti jednog vremenskog<br />niza, još uvek nije dostigla njenu reputaciju. Cilj ove disertacije je<br />da identifikuje probleme koji otežavaju širu primenu unakrsne<br />entropije i da predloži skup rešenja. Validacija rezultata je rađena<br />na kardiovaskularnim signalima, sistolnog krvnog pritiska i palsnog<br />inetervala snimljenim na laboratorijskim pacovima i na signalima<br />zdravih volontera.</p> / <p>Cross-approximate entropy (XApEn) quantifies a mutual orderliness of two<br />simultaneously recorded time series. Although derived from the firmly<br />establishe solitary entropies, it has never reached their reputation and<br />deployment. The aim of this thesis is to identify the problems that preclude<br />wider XApEn implementation and to develop a set of solutions. Results were<br />validated using the cardiovascular time series, systolic blood pressure and<br />pulse interval, recorded from laboratories animals and also signals recorded<br />from healthy human volunteers.</p>

The abuse of and dependence on opioids are major public health problems, and have been the focus of intense media coverage and scholarly inquiry. This research explores the problem in Iowa through the lens of infectious disease modeling. We wanted to identify the current state of the crisis, factors affecting the progression of the addiction process, and evaluate interventions as data becomes available. We introduced a novel sequential Approximate Bayesian Computation technique to address shortcomings of existing methods in this complex problem space, after surveying the literature for available Bayesian computation techniques. A spatial compartmental model was used which allowed forward and backward progression through susceptible, exposed, addicted, and removed disease states. Data for this model were compiled over the years 2006-2016 for Iowa counties, from a variety of sources. Prescription overdose deaths and treatment data were obtained from the Iowa Department of Public Health, possession and distribution arrest data were acquired from the Iowa Department of Public Safety, a measure of total available pain reliever prescriptions was derived from private health insurance claims data, and population totals were obtained from the US Census Bureau. Inference was conducted in a Bayesian framework. A measure called the empirically adjusted reproductive number which estimates the expected number of new users generated from a single user was used to examine the growth of the crisis. Results expose the trend in recruitment of new users, and peak recruitment times. While we identify an overall decrease in the rate of spread during the study period, the scope of the problem remains severe, and interesting outlying trends require further investigation. In addition, an examination of the reproductive numbers estimated for contact within and between counties indicates that medical exposure, rather than spread through social networks, may be the key driver of this crisis.

Let 𝑋 be a compact subset of the complex plane and denote by 𝑅𝑝(𝑋) the closure of rational functions with poles off 𝑋 in the 𝐿𝑝(𝑋) norm. We show that if a point 𝑥0 admits a bounded point derivation on 𝑅𝑝(𝑋) for 𝑝 > 2, then there is an approximate derivative at 𝑥0. We also prove a similar result for higher order bounded point derivations. This extends a result of Wang, which was proven for 𝑅(𝑋), the uniform closure of rational functions with poles off 𝑋. In addition, we show that if a point 𝑥0 admits a bounded point derivation on 𝑅(𝑋) and if 𝑋 contains an interior cone, then the bounded point derivation can be represented by the difference quotient if the limit is taken over a non-tangential ray to 𝑥0. We also extend this result to the case of higher order bounded point derivations. These results were first shown by O'Farrell; however, we prove them constructively by explicitly using the Cauchy integral formula.

Data-driven decision making has become more popular in today’s businesses including logistics and vehicle routing. Leveraging historical data, companies can achieve goals such as customer satisfaction management, scalable and efficient operation, and higher overall revenue. In the management of customer satisfaction, logistics companies use consistent assignment of their drivers to customers over time. Creating this consistency takes time and depends on the history experienced between the company and the customer. While pursuing this goal, companies trade off the cost of capacity with consistency because demand is unknown on a daily basis. We propose concepts and methods that enable a parcel delivery company to balance the trade-off between cost and customer satisfaction. We use clustering methods that use cumulative historical service data to generate better consistency using the information entropy measure. Parcel delivery companies route many vehicles to serve customer requests on a daily basis. While clustering was important to the development of early routing algorithms, modern solution methods rely on metaheuristics, which are not easily deployable and often do not have open source code bases. We propose a two-stage, shape-based clustering approach that efficiently obtains a clustering of delivery request locations. Our solution technique is based on creating clusters that form certain shapes with respect to the depot. We obtain a routing solution by ordering all locations in every cluster separately. Our results are competitive with a state-of-the-art vehicle routing solver in terms of quality. Moreover, the results show that the algorithm is more scalable and is robust to problem parameters in terms of runtime. Fish trawling can be considered as a vehicle routing problem where the main objective is to maximize the amount of fish (revenue) facing uncertainty on catch. This uncertainty creates an embedded prediction problem before deciding where to harvest. Using previous catch data to train prediction models, we solve the routing problem a fish trawler faces using dynamically updated routing decisions allowing for spatiotemporal correlation in the random catch. We investigate the relationship between the quality of predictions and the quality of revenue generated as a result.

In this paper, we develop a novel index structure to support efficient approximate k-nearest neighbor (KNN) query in high-dimensional databases. In high-dimensional spaces, the computational cost of the distance (e.g., Euclidean distance) between two points contributes a dominant portion of the overall query response time for memory processing. To reduce the distance computation, we first propose a structure (BID) using BIt-Difference to answer approximate KNN query. The BID employs one bit to represent each feature vector of point and the number of bit-difference is used to prune the further points. To facilitate real dataset which is typically skewed, we enhance the BID mechanism with clustering, cluster adapted bitcoder and dimensional weight, named the BID⁺. Extensive experiments are conducted to show that our proposed method yields significant performance advantages over the existing index structures on both real life and synthetic high-dimensional datasets. / Singapore-MIT Alliance (SMA)

Boundary-contact problems (BCPs) are studied within the frames of classical mathematical theory of elasticity and plasticity elaborated by Landau, Kupradze, Timoshenko, Goodier, Fichera and many others on the basis of analysis of two- and three-dimensional boundary value problems for linear partial differential equations. A great attention is traditionally paid both to theoretical investigations using variational methods and boundary singular integral equations (Muskhelishvili) and construction of solutions in the form that admit efficient numerical evaluation (Kupradze). A special family of BCPs considered by Shtaerman, Vorovich, Alblas, Nowell, and others arises within the frames of the models of squeezing thin multilayer elastic sheets. We show that mathematical models based on the analysis of BCPs can be also applied to modeling of the clich\'-surface printing contacts and paper surface compressibility in flexographic printing. The main result of this work is formulation and complete investigation of BCPs in layered structures, which includes both the theoretical (statement of the problems, solvability and uniqueness) and applied parts (approximate and numerical solutions, codes, simulation). We elaborate a mathematical model of squeezing a thin elastic sheet placed on a stiff base without friction by weak loads through several openings on one of its boundary surfaces. We formulate and consider the corresponding BCPs in two- and three-dimensional bands, prove the existence and uniqueness of solutions, and investigate their smoothness including the behavior at infinity and in the vicinity of critical points. The BCP in a two-dimensional band is reduced to a Fredholm integral equation (IE) with a logarithmic singularity of the kernel. The theory of logarithmic IEs developed in the study includes the analysis of solvability and development of solution techniques when the set of integration consists of several intervals. The IE associated with the BCP is solved by three methods based on the use of Fourier-Chebyshev series, matrix-algebraic determination of the entries in the resulting infinite system matrix, and semi-inversion. An asymptotic theory for the BCP is developed and the solutions are obtained as asymptotic series in powers of the characteristic small parameter. We propose and justify a technique for the solution of BCPs and boundary value problems with boundary conditions of mixed type called the approximate decomposition method (ADM). The main idea of ADM is simplifying general BCPs and reducing them to a chain of auxiliary problems for 'shifted' Laplacian in long rectangles or parallelepipeds and then to a sequence of iterative problems such that each of them can be solved (explicitly) by the Fourier method. The solution to the initial BCP is then obtained as a limit using a contraction operator, which constitutes in particular an independent proof of the BCP unique solvability. We elaborate a numerical method and algorithms based on the approximate decomposition and the computer codes and perform comprehensive numerical analysis of the BCPs including the simulation for problems of practical interest. A variety of computational results are presented and discussed which form the basis for further applications for the modeling and simulation of printing-plate contact systems and other structures of flexographic printing. A comparison with finite-element solution is performed.

A two-dimensional thin-layer Navier-Stokes cascade flow solver for turbomachinery is developed. A second-order finite-difference scheme and a second and fourth-difference dissipation scheme are used. Periodic and non-reflecting inlet and outlet boundary conditions are implemented into the approximate-factorization numerical method. Turbulence is modeled through the one-equation Spalart-Allmaras model. A two-dimensional turbomachinery cascade structured grid generator is developed to produce six-block H-type grids. The validity of this work is tested in various ways. A grid convergence study is performed showing the effect of grid density. The non-reflecting inlet and outlet boundary conditions are tested for boundary placement influence. Comparisons of the flow solver numerical results are performed against experimental results. A Mach number sweep and angle of attack sweep are performed on two similar transonic turbine cascades.

A two-dimensional thin-layer Navier-Stokes cascade flow solver for turbomachinery is developed. A second-order finite-difference scheme and a second and fourth-difference dissipation scheme are used. Periodic and non-reflecting inlet and outlet boundary conditions are implemented into the approximate-factorization numerical method. Turbulence is modeled through the one-equation Spalart-Allmaras model. A two-dimensional turbomachinery cascade structured grid generator is developed to produce six-block H-type grids. The validity of this work is tested in various ways. A grid convergence study is performed showing the effect of grid density. The non-reflecting inlet and outlet boundary conditions are tested for boundary placement influence. Comparisons of the flow solver numerical results are performed against experimental results. A Mach number sweep and angle of attack sweep are performed on two similar transonic turbine cascades.