Computer Engineering Commons^™

Query Expansion Techniques For Enterprise Search, Eric M. Domke

Masters Theses

Although web search remains an active research area, interest in enterprise search has waned. This is despite the fact that the market for enterprise search applications is expected to triple within the next six years, and that knowledge workers spend an average of 1.6 to 2.5 hours each day searching for information. To improve search relevancy, and hence reduce this time, an enterprise- focused application must be able to handle the unique queries and constraints of the enterprise environment. The goal of this thesis research was to develop, implement, and study query expansion techniques that are most effective at improving …

Automated Program Profiling And Analysis For Managing Heterogeneous Memory Systems, Adam Palmer Howard

Masters Theses

Many promising memory technologies, such as non-volatile, storage-class memories and high-bandwidth, on-chip RAMs, are beginning to emerge. Since each of these new technologies present tradeoffs distinct from conventional DRAMs, next-generation systems are likely to include multiple tiers of memory storage, each with their own type of devices. To efficiently utilize the available hardware, such systems will need to alter their data management strategies to consider the performance and capabilities provided by each tier.

This work explores a variety of cross-layer strategies for managing application data in heterogeneous memory systems. We propose different program profiling-based techniques to automatically partition program allocation …

Analyzing Spark Performance On Spot Instances, Jiannan Tian

Masters Theses

Amazon Spot Instances provide inexpensive service for high-performance computing. With spot instances, it is possible to get at most 90% off as discount in costs by bidding spare Amazon Elastic Computer Cloud (Amazon EC2) instances. In exchange for low cost, spot instances bring the reduced reliability onto the computing environment, because this kind of instance could be revoked abruptly by the providers due to supply and demand, and higher-priority customers are first served.

To achieve high performance on instances with compromised reliability, Spark is applied to run jobs. In this thesis, a wide set of spark experiments are conducted to …

Navigation Instruction Validation Tool And Indoor Wayfinding Training System For People With Disabilities, Linlin Ding

Masters Theses

According to World Health Survey, there are 785 million (15.6%) people in the world that live with a disability. It is a well-known fact that lack of access to public transportation is a barrier for people with disabilities in seeking work or accessing health care. In this research, we seek to increase access to public transportation by introducing a virtual pre-travel training system that enables people with disabilities to get familiar with a public transportation venue prior to arriving at the venue. Using this system, users establish a mental map of the target environment prior to their arrival to the …

Efficient Scaling Of A Web Proxy Cluster, Hao Zhang

Masters Theses

With the continuing growth in network traffic and increasing diversity in web content, web caching, together with various network functions (NFs), has been introduced to enhance security, optimize network performance, and save expenses. In a large enterprise network with more than tens of thousands of users, a single proxy server is not enough to handle a large number of requests and turns to group processing. When multiple web cache proxies are working as a cluster, they talk with each other and share cached objects by using internet cache protocol (ICP). This leads to poor scalability.

This thesis describes the development …

Query On Knowledge Graphs With Hierarchical Relationships, Kaihua Liu

Masters Theses

The dramatic popularity of graph database has resulted in a growing interest in graph queries. Two major topics are included in graph queries. One is based on structural relationship to find meaningful results, such as subgraph pattern match and shortest-path query. The other one focuses on semantic-based query to find question answering from knowledge bases. However, most of these queries take knowledge graphs as flat forms and use only normal relationship to mine these graphs, which may lead to mistakes in the query results. In this thesis, we find hierarchical relationship in the knowledge on their semantic relations and make …

Magneto-Electric Approximate Computational Framework For Bayesian Inference, Sourabh Kulkarni

Masters Theses

Probabilistic graphical models like Bayesian Networks (BNs) are powerful artificial-intelligence formalisms, with similarities to cognition and higher order reasoning in the human brain. These models have been, to great success, applied to several challenging real-world applications. Use of these formalisms to a greater set of applications is impeded by the limitations of the currently used software-based implementations. New emerging-technology based circuit paradigms which leverage physical equivalence, i.e., operating directly on probabilities vs. introducing layers of abstraction, promise orders of magnitude increase in performance and efficiency of BN implementations, enabling networks with millions of random variables. While majority of applications with …

Oracle Guided Incremental Sat Solving To Reverse Engineer Camouflaged Circuits, Xiangyu Zhang

Masters Theses

This study comprises two tasks. The first is to implement gate-level circuit camouflage techniques. The second is to implement the Oracle-guided incremental de-camouflage algorithm and apply it to the camouflaged designs.

The circuit camouflage algorithms are implemented in Python, and the Oracle- guided incremental de-camouflage algorithm is implemented in C++. During this study, I evaluate the Oracle-guided de-camouflage tool (Solver, in short) performance by de-obfuscating the ISCAS-85 combinational benchmarks, which are camouflaged by the camouflage algorithms. The results show that Solver is able to efficiently de-obfuscate the ISCAS-85 benchmarks regardless of camouflaging style, and is able to do so 10.5x …

Tiled Danna: Dynamic Adaptive Neural Network Array Scaled Across Multiple Chips, Patricia Jean Eckhart

Masters Theses

Tiled Dynamic Adaptive Neural Network Array(Tiled DANNA) is a recurrent spiking neural network structure composed of programmable biologically inspired neurons and synapses that scales across multiple FPGA chips. Fire events that occur on and within DANNA initiate spiking behaviors in the programmable elements allowing DANNA to hold memory through the synaptic charge propagation and neuronal charge accumulation. DANNA is a fully digital neuromorphic computing structure based on the NIDA architecture. To support initial prototyping and testing of the Tiled DANNA, multiple Xilinx Virtex 7 690Ts were leveraged. The primary goal of Tiled DANNA is to support scaling of DANNA neural …

Scalable High-Speed Communications For Neuromorphic Systems, Aaron Reed Young

Masters Theses

Field-programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), and other chip/multi-chip level implementations can be used to implement Dynamic Adaptive Neural Network Arrays (DANNA). In some applications, DANNA interfaces with a traditional computing system to provide neural network configuration information, provide network input, process network outputs, and monitor the state of the network. The present host-to-DANNA network communication setup uses a Cypress USB 3.0 peripheral controller (FX3) to enable host-to-array communication over USB 3.0. This communications setup has to run commands in batches and does not have enough bandwidth to meet the maximum throughput requirements of the DANNA device, resulting …

Virtualization Of Closed-Loop Sensor Networks, Priyanka Dattatri Kedalagudde

Masters Theses

The existing closed-loop sensor networks are based on architectures that are designed and implemented for one specific application and require dedicated sensing and computational resources. This prevents the sharing of these networks. In this work, we propose an architecture of virtualization to allow sharing of closed-loop sensor networks. We also propose a scheduling approach that will manage requests from competing applications and evaluate their impact on system utilization against utilization achieved by more traditional, dedicated sensor networks. These algorithms are evaluated through trace-driven simulations, where the trace data is taken from CASA’s closed-loop weather radar sensor network. Results from this …

Optimization Of Spatial Convolution In Convnets On Intel Knl, Sangamesh Nagashattappa Ragate

Masters Theses

Most of the experts admit that the true behavior of the neural network is hard to predict. It is quite impossible to deterministically prove the working of the neural network as the architecture gets bigger, yet, it is observed that it is possible to apply a well engineered network to solve one of the most abstract problems like image recognition with substantial accuracy. It requires enormous amount of training of a considerably big and complex neural network to understand its behavior and iteratively improve its accuracy in solving a certain problem. Deep Neural Networks, which are fairly popular nowadays deal …

Explorations Into Machine Learning Techniques For Precipitation Nowcasting, Aditya Nagarajan

Masters Theses

Recent advances in cloud-based big-data technologies now makes data driven solutions feasible for increasing numbers of scientific computing applications. One such data driven solution approach is machine learning where patterns in large data sets are brought to the surface by finding complex mathematical relationships within the data. Nowcasting or short-term prediction of rainfall in a given region is an important problem in meteorology. In this thesis we explore the nowcasting problem through a data driven approach by formulating it as a machine learning problem.

State-of-the-art nowcasting systems today are based on numerical models which describe the physical processes leading to …

Achieving Perfect Location Privacy In Wireless Devices Using Anonymization, Zarrin Montazeri

Masters Theses

The popularity of mobile devices and location-based services (LBS) have created great concerns regarding the location privacy of the users of such devices and services. Anonymization is a common technique that is often being used to protect the location privacy of LBS users. This technique assigns a random pseudonym to each user and these pseudonyms can change over time. Here, we provide a general information theoretic definition for perfect location privacy and prove that perfect location privacy is achievable for mobile devices when using the anonymization technique appropriately. First, we assume that the user’s current location is independent from her …

Analysis Of Outsourcing Data To The Cloud Using Autonomous Key Generation, Mortada Abdulwahed Aman

Masters Theses

"Cloud computing, a technology that enables users to store and manage their data at a low cost and high availability, has been emerging for the past few decades because of the many services it provides. One of the many services cloud computing provides to its users is data storage. The majority of the users of this service are still concerned to outsource their data due to the integrity and confidentiality issues, as well as performance and cost issues, that come along with it. These issues make it necessary to encrypt data prior to outsourcing it to the cloud. However, encrypting …

Predicting The Impact Of Data Corruption On The Operation Of Cyber-Physical Systems, Erik David Burgdorf

Masters Theses

"Cyber-physical systems, where computing and communication are used to fortify and streamline the operation of a physical infrastructure, now comprise the foundation of much of modern critical infrastructure. These systems are typically large in scale and highly interconnected, and span application domains from power and water distribution to autonomous vehicle control and collaborative robotics. Intelligent decision support in these systems is heavily reliant on the availability of sufficient and sufficiently correct data. Failure or malfunction of these systems can have devastating consequences in terms of public safety, financial losses, or both.

The research described in this thesis aims to predict …

Novel Approaches For Efficient Stochastic Computing, Ramu Seva

Masters Theses

"This thesis is comprised of two papers, where the first paper presents a novel approach for parallel implementation of SC using FPGA (Field Programmable Gate Array). This paper makes use of the distributed memory elements of FPGAs (i.e., look-up-tables -LUTs) to achieve this. An attempt has been made to build the stochastic number generators (SNGs) by using the proposed LUT approach. The construction of these SNGs has been influenced by the Quasi-random number sequences, which provide the advantage of reducing the random fluctuations present in the pseudo-random number generators such as LFSR (Linear Feedback Shift Register) as well as the …