Computer Engineering Commons^™

Infrastructure For Performance Tuning Mpi Applications, Kathryn Marie Mohror

Dissertations and Theses

Clusters of workstations are becoming increasingly popular as a low-budget alternative for supercomputing power. In these systems, message-passing is often used to allow the separate nodes to act as a single computing machine. Programmers of such systems face a daunting challenge in understanding the performance bottlenecks of their applications. This is largely due to the vast amount of performance data that is collected, and the time and expertise necessary to use traditional parallel performance tools to analyze that data.

The goal of this project is to increase the level of performance tool support for message-passing application programmers on clusters of …

Minimality Of Solution Update In Conflict Resolution: An Application Of Revision Programming To Von Neumann-Morgenstern Approach, Inna Pivkina, Vladik Kreinovich

Departmental Technical Reports (CS)

In a 1944 book that started game theory (and mathematical approach to conflict resolution), von Neumann and Morgenstern proposed the notion of a solution. When the situation changes, the old solution is often no longer a solution, so it needs to be updated. In practical applications, it is usually desirable to keep the solution change "minimal" in some reasonable sense. We show that for a seemingly straightforward formalization of this minimality, checking whether a change is minimal is NP-hard. We also show that by representing the notion of a solution as a collection of revision rules, we can produce a …

Turning An Articulated Truck On A Spreadsheet, Jim Mcgovern

Articles

This paper describes how a standard office spreadsheet application was used to set up an engineering model for the simulation and analysis of the turning of articulated trucks

Vector Dissipativity Theory For Discrete-Time Large-Scale Nonlinear Dynamical Systems, Wassim M. Haddad, Qing Hui, Vijaysekhar Chellaboina, Sergey Nersesov

Department of Electrical and Computer Engineering: Faculty Publications

In analyzing large-scale systems, it is often desirable to treat the overall system as a collection of interconnected subsystems. Solution properties of the large-scale system are then deduced from the solution properties of the individual subsystems and the nature of the system interconnections. In this paper, we develop an analysis framework for discrete-time large-scale dynamical systems based on vector dissipativity notions. Specifically, using vector storage functions and vector supply rates, dissipativity properties of the discrete-time composite large-scale system are shown to be determined from the dissipativity properties of the subsystems and their interconnections. In particular, extended Kalman-Yakubovich-Popov conditions, in terms …

Storage Coalescing, Delvin C. Defoe

All Computer Science and Engineering Research

Typically, when a program executes, it creates objects dynamically and requests storage for its objects from the underlying storage allocator. The patterns of such requests can potentially lead to internal fragmentation as well as external fragmentation. Internal fragmentation occurs when the storage allocator allocates a contiguous block of storage to a program, but the program uses only a fraction of that block to satisfy a request. The unused portion of that block is wasted since the allocator cannot use it to satisfy a subsequent allocation request. External fragmentation, on the other hand, concerns chunks of memory that reside between allocated …

Wright State University College Of Engineering And Computer Science Bits And Pcs Newsletter, Volume 20, Number 2, October 2003, College Of Engineering And Computer Science, Wright State University

BITs and PCs Newsletter

An eight page newsletter created by the Wright State University College of Engineering and Computer Science that addresses the current affairs of the college.

Sensitivity Analysis Of Neural Control, Chin-Wang Tao, Hung T. Nguyen, J. T. Yao, Vladik Kreinovich

Departmental Technical Reports (CS)

We provide explicit formulas that describe how sensitive the resulting signal of a neural network is to the measurement errors with which we measure the inputs.

Real-Time Algorithms For Statistical Analysis Of Interval Data, Berlin Wu, Hung T. Nguyen, Vladik Kreinovich

Departmental Technical Reports (CS)

When we have only interval ranges [xi] of sample values x1,...,xn, what is the interval [V] of possible values for the variance V of these values? There are quadratic time algorithms for computing the exact lower bound V- on the variance of interval data, and for computing V+ under reasonable easily verifiable conditions. The problem is that in real life, we often make additional measurements. In traditional statistics, if we have a new measurement result, we can modify the value of variance in constant time. In contrast, previously known algorithms for processing interval data required that, once a new data …

A New Differential Formalism For Interval-Valued Functions And Its Potential Use In Detecting 1-D Landscape Features, Vladik Kreinovich, Hung T. Nguyen, Gracaliz Pereira Dimuro, Antonio Carlos Da Rocha Costa, Benjamin Rene Callejas Bedregal

Departmental Technical Reports (CS)

In many practical problems, it is important to know the slope (derivative) dy/dx of one quantity y with respect to some other quantity x. For example, different 1-D landscape features can be characterized by different values of the derivative dy/dx, where y is an altitude, and x is a horizontal coordinate. In practice, we often know the values of y(x) for different x with interval uncertainty. How can we then find the set of possible values of the slope? In this paper, we formulate this problem of differentiating interval-values functions in precise terms, and we describe an (asymptotically) optimal algorithm …

Fast Multiplication Of Interval Matrices (Interval Version Of Strassen's Algorithm), Martine Ceberio, Vladik Kreinovich

Departmental Technical Reports (CS)

Strassen's algorithm multiplies two numerical matrices fast, but when applied to interval matrices, leads to excess width. We use Rump's interval arithmetic to propose an interval version of Strassen's algorithm whose only excess width is in second order terms.

Separating Components In Interval-Valued Images, Marilton Sanchotene De Aguiar, Gracaliz Pereira Dimuro, Antonio Carlos Da Rocha Costa, Andrei Finkelstein, Vladik Kreinovich

Departmental Technical Reports (CS)

In many applications of imaging, we would like to know whether we have an image of a single-component object or an image of an object that consists of several components. Many algorithms have been designed to solve this problem; however, these algorithms are all heuristic. Often, according to some reasonable methods, we have a single component, while according to some other equally reasonable methods, the same image have multiple components. It is desirable to produce reliable methods, so that if a method claims that there are multiple components, then it should mean that the observed data is incompatible with the …

Novel Approaches To Numerical Software With Result Verification, Laurent Granvilliers, Vladik Kreinovich, Norbert Mueller

Departmental Technical Reports (CS)

Traditional design of numerical software with result verification is based on the assumption that we know the algorithm f(x_1,...,xn) that transforms input x1,...,xn into the output y=f(x1,...,xn), and we know the intervals of possible values of the inputs. Many real-life problems go beyond this paradigm. In some cases, we do not have an algorithm f, we only know some relation (constraints) between xi and y. In other cases, in addition to knowing the intervals [xi], we may know some relations between xi; we may have some information about the probabilities of different values of xi, and we may know the …

Greedy Algorithms For Optimizing Multivariate Horner Schemes, Martine Ceberio, Vladik Kreinovich

Departmental Technical Reports (CS)

For univariate polynomials f(x1), Horner scheme provides the fastest way to compute the value. For multivariate polynomials, several different version of Horner scheme are possible; it is not clear which of them is optimal. In this paper, we propose a greedy algorithm that will hopefully lead to good computation times.

A univariate Horner scheme has another advantage: if the value x1 is known with uncertainty, and we are interested in the resulting uncertainty in f(x1), then Horner scheme leads to a better estimate for this uncertainty than many other ways of computing f(x1). The second greedy algorithm that we propose …

Interval Approach To Phase Measurements Can Lead To Arbitrarily Complex Sets - A Theorem And Ways Around It, Bharat C. Mulupuru, Vladik Kreinovich, Roberto Osegueda

Departmental Technical Reports (CS)

We are often interested in phases of complex quantities; e.g., in non-destructive testing of aerospace structures, important information comes from phases of Eddy current and magnetic resonance.

For each measurement, we have an upper bound D on the measurement error dx=X-x, so when the measurement result is X, we know that the actual value x is in [X-D,X+D]. Often, we have no information about probabilities of different values, so this interval is our only information about x. When the accuracy is not sufficient, we perform several repeated measurements, and conclude that x belongs to the intersection of the corresponding intervals. …

2003 (Fall) Ensi Informer Magazine, Morehead State University. Engineering Sciences Department

ENSI Informer Magazine Archive

The ENSI Informer Magazine published in the fall of 2003.

Unidirectional Cordic For Efficient Computation Of Trigonometric And Hyperbolic Functions, Satish Ravichandran

Electrical & Computer Engineering Theses & Dissertations

CORDIC (Coordinate Rotation Digital Computer) is an iterative algorithm to compute values of trigonometric, logarithmic and transcendental functions by performing vector rotations, which can be implemented with only shift and add operations in a digital system. CORDIC algorithms are extensively used in the areas of digital signal processing, digital image processing and artificial neural networks. A new technique, named unidirectional CORDIC, for efficient computation of trigonometric and hyperbolic functions is presented in this thesis. In the conventional CORDIC algorithm, the vector rotations are performed in both clockwise and counterclockwise directions, but in the unidirectional CORDIC the vectors are rotated only …

Examination Of Focused Beam Propagation Through A Finite Non-Reciprocal Planar Chiral Slab Using Complex Fresnel Coefficients And Dual Transforms, Monish Ranjan Chatterjee, Sumit Nema, Partha P. Banerjee

Electrical and Computer Engineering Faculty Publications

Recently, uniform plane wave propagation across a nonchiral-chiral interface was examined via the development of appropriate Fresnel coefficients. In this extension, propagation of focused uniform and profiled beams through a finite, planar nonreciprocal chiral slab is investigated using complex Fresnel coefficients via angular plane wave decomposition and dual transforms.

Cs 790-01: Introduction To Parallel Algorithms, Natsuhiko Futamura

Computer Science & Engineering Syllabi

Low-cost parallel computers such as PC clusters are becoming available, and many previously unsolvable problems can be solved using such computers. However, designing algorithms that perform well on parallel computers is often challenging. The focus of this course is on learning how to design algorithms for parallel computers and how to evaluate them.

The Effect Of Pruning And Compression On Graphical Representations Of The Output Of A Speech Recognizer, Yang Liu, Mary P. Harper, Michael T. Johnson, Leah H. Jamieson

Electrical and Computer Engineering Faculty Research and Publications

Large vocabulary continuous speech recognition can benefit from an efficient data structure for representing a large number of acoustic hypotheses compactly. Word graphs or lattices have been chosen as such an efficient interface between acoustic recognition engines and subsequent language processing modules. This paper first investigates the effect of pruning during acoustic decoding on the quality of word lattices and shows that by combining different pruning options (at the model level and word level), we can obtain word lattices with comparable accuracy to the original lattices and a manageable size. In order to use the word lattices as the input …

Wright State University College Of Engineering And Computer Science Bits And Pcs Newsletter, Volume 20, Number 1, September 2003, College Of Engineering And Computer Science, Wright State University

BITs and PCs Newsletter

An eight page newsletter created by the Wright State University College of Engineering and Computer Science that addresses the current affairs of the college.

Interval Arithmetic, Affine Arithmetic, Taylor Series Methods: Why, What Next?, Nedialko S. Nedialkov, Vladik Kreinovich, Scott A. Starks

Departmental Technical Reports (CS)

In interval computations, the range of each intermediate result r is described by an interval [r]. To decrease excess interval width, we can keep some information on how r depends on the input x=(x1,...,xn). There are several successful methods of approximating this dependence; in these methods, the dependence is approximated by linear functions (affine arithmetic) or by general polynomials (Taylor series methods). Why linear functions and polynomials? What other classes can we try? These questions are answered in this paper.

A Full Function-Based Calculus Of Directed And Undirected Intervals: Markov's Interval Arithmetic Revisited, Juergen Wolff Von Gudenberg, Vladik Kreinovich

Departmental Technical Reports (CS)

This paper proposes a new interpretation of intervals as classes of functions having the same domain. Interval operations are seen as operations on these classes. This approach allows to recover Markov's directed interval arithmetic by taking into account the monotonicity of the functions.

A Feasible Algorithm For Locating Concave And Convex Zones Of Interval Data And Its Use In Statistics-Based Clustering, Vladik Kreinovich, Eric J. Pauwels, Scott Ferson, Lev Ginzburg

Departmental Technical Reports (CS)

Often, we need to divide n objects into clusters based on the value of a certain quantity x. For example, we can classify insects in the cotton field into groups based on their size and other geometric characteristics. Within each cluster, we usually have a unimodal distribution of x, with a probability density d(x) that increases until a certain value x0 and then decreases. It is therefore natural, based on d(x), to determine a cluster as the interval between two local minima, i.e., as a union of adjacent increasing and decreasing segments. In this paper, we describe a feasible algorithm …

A Communications Modeling System For Swarm-Based Sensors, Brian A. Kadrovach

Theses and Dissertations

Today's information age has exploded the amount of data available to decision makers at all levels of the control hierarchy. The miniaturization and proliferation of sensor technology has enabled extensive detection and monitoring and advances in computational capabilities have provided for embedded data analysis and the generation of information from raw data. Additionally, with the miniaturization of mechanical systems it is possible to provide platforms for sensor suites that are capable of mobility and limited autonomy. Swarming or bio-emergent behavior problems, provides a robust scalable mechanism for organizing large numbers of mobile sensor platforms. However, the mobility dynamics of swarm …

Toward Building Conversational Spoken-Language Interfaces: Acknowledgment Use In American English And Mexican Spanish, Karen Ward, Tasha Hollingsed, Javier A. Aldaz Salmon

Graduate Student Papers (CS)

Should spoken-language interfaces incorporate human discourse phenomena? Acknowledgments, for example, are ubiquitous in human conversation but are rare in human-computer interaction. Are people unwilling to use this human convention when talking to a machine, or is their scarcity due to the design of current spoken-language interfaces? We found that, given a simple spoken-language interface that responded to acknowledgments, over two thirds of subjects used acknowledgments at least once, about the same number that used more traditional commands to control the interface. These results were consistent for both Mexican Spanish and American English versions of the interface, and they suggest that …

Ip Traceback With Deterministic Packet Marking Dpm, Andrey Belenky

Dissertations

In this dissertation, a novel approach to Internet Protocol (IP) Traceback - Deterministic Packet Marking (DPM) is presented. The proposed approach is scalable, simple to implement, and introduces no bandwidth and practically no processing overhead on the network equipment. It is capable of tracing thousands of simultaneous attackers during a Distributed Denial of Service (DDoS) attack. Given sufficient deployment on the Internet, DPM is capable of tracing back to the slaves for DDoS attacks which involve reflectors. Most of the processing is done at the victim. The traceback process can be performed post-mortem, which allows for tracing the attacks that …

Optimal And Intelligent Decision Making In Sustainable Development Of Electronic Products, Meimei Gao

Dissertations

Increasing global population and consumption are causing declining natural and social systems. Multi-lifecycle engineering and sustainable development address these issues by integrating strategies for economic successes, environmental quality, and social equity. Based on multi-lifecycle engineering and sustainable development concepts, this doctoral dissertation aims to provide decision making approaches to growing a strong industrial economy while maintaining a clean, healthy environment. The research develops a methodology to complete both the disassembly leveling and bin assignment decisions in demanufacturing through balancing the disassembly efforts, value returns, and environmental impacts. The proposed method is successfully implemented into a demanufacturing module of a Multi-LifeCycle …

Internet Traffic Over Wide Area Network - Statistical Modeling And Analysis, Yan Yi Lee

Theses

This thesis provides results on trace data collecting and modeling of the Internet traffic over a Wide Area Network. The study is based on extensive data, gathered by tracing the actual packet exchange at the interfaces of devices on the WAN of BCCLS (Bergen County Cooperative Library System). A powerful network monitoring system WhatsUp Gold was employed to monitor the WAN, and logged the byte rates and packet rates sent and received at the interfaces. Different statistical distributions were employed to model the data traces. In comparison with some other statistical models - normal, lognormal, Weibull, and Pareto, which are …

Low Interference Routing For Wireless Ad-Hoc Networks, Mohit Gupta

Theses

In this thesis the primary focus is on the problem of interference between messages. The thesis discusses why the messages are blocked in a system? How adding a message impacts the cost of all other available links, which can be established in the system.

This thesis analyzes how the availability of channels, increase in number of nodes and increase in the transmission range help in increasing the number of messages that can be handled in the network. It is also analyzes how critical is the selection of the maximum transmission range MTR, transmission range TR and required transmission range RTR. …

Design And Evaluation Of Processes For Fuel Fabrication: Quarterly Progress Report #8, Georg F. Mauer

Fuels Campaign (TRP)

The eighth quarter of the project covered the following:

· Mr. Richard Silva continued the development of a simulation model with a Waelischmiller hot cell robot. Rich will continue to develop detailed 3-D process simulation models as his M.Sc. thesis project.

· A paper submitted for ANS for the Winter Annual Meeting on hot cell robotics was accepted for presentation and publication.

· Further advances on Concepts and Methods for Vision-Based Hot Cell Supervision and control, focusing on rule-based object recognition (Ph.D. Student Jae-Kyu Lee)

· Undergraduate student Jamil Renno created simulations of fuel pin assembly (robotic insertion of pellets …