Systems Architecture Commons^™

Xtreme-Noc: Extreme Gradient Boosting Based Latency Model For Network-On-Chip Architectures, Ilma Sheriff

All Graduate Theses, Dissertations, and Other Capstone Projects

Multiprocessor System-on-Chip (MPSoC) integrating heterogeneous processing elements (CPU, GPU, Accelerators, memory, I/O modules ,etc.) are the de-facto design choice to meet the ever-increasing performance/Watt requirements from modern computing machines. Although at consumer level the number of processing elements (PE) are limited to 8-16, for high end servers, the number of PEs can scale up to hundreds. A Network-on-Chip (NoC) is a microscale network that facilitates the packetized communication among the PEs in such complex computational systems. Due to the heterogeneous integration of the cores, execution of diverse (serial and parallel) applications on the PEs, application mapping strategies, and many other …

Occam Software (And Manual) For Reconstructability Analysis, Martin Zwick, Kenneth Willett, Joe Fusion, Heather Alexander

Complex Systems Faculty Publications and Presentations

OCCAM is a Discrete Multivariate Modeling (DMM) tool based on the methodology of Reconstructability Analysis (RA). As an acronym it stands for Organizational Complexity Computation and Modeling, and the name is also a reference (with a non-standard spelling) to Ockham’s Razor. The principal programmers of its current version have been Kenneth Willett, Joe Fusion, and Heather Alexander. Ken Willett totally rewrote earlier versions of OCCAM. His version was originally called “OCCAM3” to distinguish it from these earlier OCCAM incarnations; the “3” has finally been dropped.

OCCAM’s typical use is to analyze data involving a large number of discrete variables, but …

On Studying Distributed Machine Learning, Simeon Eberz

Senior Honors Theses

The Internet of Things (IoT) is utilizing Deep Learning (DL) for applications such as voice or image recognition. Processing data for DL directly on IoT edge devices reduces latency and increases privacy. To overcome the resource constraints of IoT edge devices, the computation for DL inference is distributed between a cluster of several devices. This paper explores DL, IoT networks, and a novel framework for distributed processing of DL in IoT clusters. The aim is to facilitate and simplify deployment, testing, and study of a distributed DL system, even without physical devices. The contributions of this paper are a deployment …

Efficient Modeling Of Random Sampling-Based Lru Cache, Junyao Yang

Dissertations, Master's Theses and Master's Reports

The Miss Ratio Curve (MRC) is an important metric and effective tool for caching system performance prediction and optimization. Since the Least Recently Used (LRU) replacement policy is the de facto policy for many existing caching systems, most previous studies on efficient MRC construction are predominantly focused on the LRU replacement policy. Recently, the random sampling-based replacement mechanism, as opposed to replacement relying on the rigid LRU data structure, gains more popularity due to its lightweight and flexibility. To approximate LRU, at replacement times, the system randomly selects K objects and replaces the least recently used object among the sample. …

Secure Network Access Via Ldap, Nicholas Valaitis

Williams Honors College, Honors Research Projects

Networks need the ability to be access by secure accounts and users. The goal of this project is to configure and expand on LDAP configurations with considerations for AAA via TACACS+ and Radius for network equipment. This will provide adequate security for any given network in terms of access and prevent lose of access to devices which happens all to often with locally configured accounts on devices.

Image Spam Classification With Deep Neural Networks, Ajay Pal Singh, Katerina Potika

Faculty Publications, Computer Science

Image classification is a fundamental problem of computer vision and pattern recognition. We focus on images that contain spam. Spam is unwanted bulk content, and image spam is unwanted content embedded inside the images. Image spam potentially creates a threat to the credibility of any email-based communication system. While a lot of machine learning techniques are successful in detecting textual based spam, this is not the case for image spams, which can easily evade these textual-spam detection systems. In our work, we explore and evaluate four deep learning techniques that detect image spams. First, we train deep neural networks using …

Theory-Inspired Path-Regularized Differential Network Architecture Search, Pan Zhou, Caiming Xiong, Richard Socher, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

Despite its high search efficiency, differential architecture search (DARTS) often selects network architectures with dominated skip connections which lead to performance degradation. However, theoretical understandings on this issue remain absent, hindering the development of more advanced methods in a principled way. In this work, we solve this problem by theoretically analyzing the effects of various types of operations, e.g. convolution, skip connection and zero operation, to the network optimization. We prove that the architectures with more skip connections can converge faster than the other candidates, and thus are selected by DARTS. This result, for the first time, theoretically and explicitly …

Evaluation Of Reliability Indicators Of Mobile Communication System Bases, Dilmurod Davronbekov, Utkir Karimovich Matyokubov, Malika Ilkhamovna Abdullayeva

Bulletin of TUIT: Management and Communication Technologies

In this study, the reliability of mobile system base stations (BTS) is assessed by analyzing data obtained on faults in about 200 BTS over a six-month period. Five BTSs with the highest number of failures and duration of failure were selected in these BTSs. Based on the data obtained, reliability parameters were calculated and compared.

The study used Weibull’s dismissal process distribution method. The breakdown times of each BTS were sorted. In all five BTS, it was found that β(where the value of β is the approximate value obtained from the values of the smallest squares of the Weibull graph), …

A Framework For Identifying Host-Based Artifacts In Dark Web Investigations, Arica Kulm

Masters Theses & Doctoral Dissertations

The dark web is the hidden part of the internet that is not indexed by search engines and is only accessible with a specific browser like The Onion Router (Tor). Tor was originally developed as a means of secure communications and is still used worldwide for individuals seeking privacy or those wanting to circumvent restrictive regimes. The dark web has become synonymous with nefarious and illicit content which manifests itself in underground marketplaces containing illegal goods such as drugs, stolen credit cards, stolen user credentials, child pornography, and more (Kohen, 2017). Dark web marketplaces contribute both to illegal drug usage …

Enhancing A Cluster-Based Tdma Mac Protocol For Vehicle-To-Vehicle Communications, Abubakar Bello Tambawal

Student Works (2020-2029)

Vehicular Ad hoc Network technology (VANET) is one of the emerging and promising wireless technology, providing support for vehicles to communicate and share resources, (such as safety messages) through vehicle-to-vehicle (V2V) communications. Sequel to that the Time Division Multiple Access (TDMA) MAC protocol using a cluster-based topology has been proposed by the research community. Most of the existing research works focused on the cluster head (CH) election with very few addressing other critical issues, including cluster formation, efficient time slot allocation, and cluster maintenance. These challenges result in an unstable cluster, which could affect the timely delivery of safety applications. …

Rethinking The Weakness Of Stream Ciphers And Its Application To Encrypted Malware Detection, William Stone, Daeyoung Kim, Victor Youdom Kemmoe, Mingon Kang, Junggab Son

School of Computing Faculty Scholarship and Creative Works

One critical vulnerability of stream ciphers is the reuse of an encryption key. Since most stream ciphers consist of only a key scheduling algorithm and an Exclusive OR (XOR) operation, an adversary may break the cipher by XORing two captured ciphertexts generated under the same key. Various cryptanalysis techniques based on this property have been introduced in order to recover plaintexts or encryption keys; in contrast, this research reinterprets the vulnerability as a method of detecting stream ciphers from the ciphertexts it generates. Patterns found in the values (characters) expressed across the bytes of a ciphertext make the ciphertext distinguishable …

Tpr: Text-Aware Preference Ranking For Recommender Systems, Yu-Neng Chuang, Chih-Ming Chen, Chuan-Ju Wang, Ming-Feng Tsai, Yuan Fang, Ee-Peng Lim

Research Collection School Of Computing and Information Systems

Textual data is common and informative auxiliary information for recommender systems. Most prior art utilizes text for rating prediction, but rare work connects it to top-N recommendation. Moreover, although advanced recommendation models capable of incorporating auxiliary information have been developed, none of these are specifically designed to model textual information, yielding a limited usage scenario for typical user-to-item recommendation. In this work, we present a framework of text-aware preference ranking (TPR) for top-N recommendation, in which we comprehensively model the joint association of user-item interaction and relations between items and associated text. Using the TPR framework, we construct a joint …

Modular Neural Networks For Low-Power Image Classification On Embedded Devices, Abhinav Goel, Sara Aghajanzadeh, Caleb Tung, Shuo-Han Chen, George K. Thiruvathukal, Yung-Hisang Lu

Computer Science: Faculty Publications and Other Works

Embedded devices are generally small, battery-powered computers with limited hardware resources. It is difficult to run deep neural networks (DNNs) on these devices, because DNNs perform millions of operations and consume significant amounts of energy. Prior research has shown that a considerable number of a DNN’s memory accesses and computation are redundant when performing tasks like image classification. To reduce this redundancy and thereby reduce the energy consumption of DNNs, we introduce the Modular Neural Network Tree architecture. Instead of using one large DNN for the classifier, this architecture uses multiple smaller DNNs (called modules) to progressively classify images …

A Statistical Impulse Response Model Based On Empirical Characterization Of Wireless Underground Channel, Abdul Salam, Mehmet C. Vuran, Suat Irmak

Faculty Publications

Wireless underground sensor networks (WUSNs) are becoming ubiquitous in many areas. The design of robust systems requires extensive understanding of the underground (UG) channel characteristics. In this paper, an UG channel impulse response is modeled and validated via extensive experiments in indoor and field testbed settings. The three distinct types of soils are selected with sand and clay contents ranging from $13\%$ to $86\%$ and $3\%$ to $32\%$, respectively. The impacts of changes in soil texture and soil moisture are investigated with more than $1,200$ measurements in a novel UG testbed that allows flexibility in soil moisture control. Moreover, the …

Direct Digital Synthesis: A Flexible Architecture For Advanced Signals Research For Future Satellite Navigation Payloads, Pranav R. Patel

Theses and Dissertations

In legacy Global Positioning System (GPS) Satellite Navigation (SatNav) payloads, the architecture does not provide the flexibility to adapt to changing circumstances and environments. GPS SatNav payloads have largely remained unchanged since the system became fully operational in April 1995. Since then, the use of GPS has become ubiquitous in our day-to-day lives. GPS availability is now a basic assumption for distributed infrastructure; it has become inextricably tied to our national power grids, cellular networks, and global financial systems. Emerging advancements of easy to use radio technologies, such as software-defined radios (SDRs), have greatly lowered the difficulty of discovery and …

An Empirical Study Of Refactorings And Technical Debt In Machine Learning Systems, Yiming Tang, Raffi Khatchadourian, Mehdi Bagherzadeh, Rhia Singh, Ajani Stewart, Anita Raja

Publications and Research

Machine Learning (ML), including Deep Learning (DL), systems, i.e., those with ML capabilities, are pervasive in today's data-driven society. Such systems are complex; they are comprised of ML models and many subsystems that support learning processes. As with other complex systems, ML systems are prone to classic technical debt issues, especially when such systems are long-lived, but they also exhibit debt specific to these systems. Unfortunately, there is a gap of knowledge in how ML systems actually evolve and are maintained. In this paper, we fill this gap by studying refactorings, i.e., source-to-source semantics-preserving program transformations, performed in real-world, open-source …

Using High-Performance Computing Profilers To Understand The Performance Of Graph Algorithms, Costain Nachuma

University of New Orleans Theses and Dissertations

An algorithm designer working with parallel computing systems should know how the characteristics of their implemented algorithm affects various performance aspects of their parallel program. It would be beneficial to these designers if each algorithm came with a specific set of standards that identified which algorithms worked better for a specified system. Therefore, the goal of this paper is to take implementations of four graphing algorithms, extract their features such as memory consumption, scalability using profilers (Vtunes /Tau) to determine which algorithms work to their fullest potential in one of the three systems: GPU, shared memory system, or distributed memory …

A Framework To Support Automatic Certification For Self-Adaptive Systems, Ioannis Nearchou

Masters Theses

Presently, cyber-physical systems are increasingly being integrated into societies, from the economic sector to the nuclear energy sector. Cyber-physical systems are systems that combine physical, digital, human, and other components, which operate through physical means and software. When system errors occur, the consequences of malfunction could negatively impact human life. Academic studies have relied on the MAPE-K feedback loop model to develop various system components to satisfy the self-adaptive features, such that violation of the safety requirements can be minimized. Assurance of system requirement satisfaction is argued through an industrial standard form, called an assurance case, which is usually applied …

Investigating Single Precision Floating General Matrix Multiply In Heterogeneous Hardware, Steven Harris

McKelvey School of Engineering Graduate Student Theses & Dissertations

The fundamental operation of matrix multiplication is ubiquitous across a myriad of disciplines. Yet, the identification of new optimizations for matrix multiplication remains relevant for emerging hardware architectures and heterogeneous systems. Frameworks such as OpenCL enable computation orchestration on existing systems, and its availability using the Intel High Level Synthesis compiler allows users to architect new designs for reconfigurable hardware using C/C++. Using the HARPv2 as a vehicle for exploration, we investigate the utility of several of the most notable matrix multiplication optimizations to better understand the performance portability of OpenCL and the implications for such optimizations on this and …

Off-Policy Reinforcement Learning For Efficient And Effective Gan Architecture Search, Tian Yuan, Wang Qin, Zhiwu Huang, Wen Li, Dengxin Dai, Minghao Yang, Jun Wang, Olga Fink

Research Collection School Of Computing and Information Systems

In this paper, we introduce a new reinforcement learning (RL) based neural architecture search (NAS) methodology for effective and efficient generative adversarial network (GAN) architecture search. The key idea is to formulate the GAN architecture search problem as a Markov decision process (MDP) for smoother architecture sampling, which enables a more effective RL-based search algorithm by targeting the potential global optimal architecture. To improve efficiency, we exploit an off-policy GAN architecture search algorithm that makes efficient use of the samples generated by previous policies. Evaluation on two standard benchmark datasets (i.e., CIFAR-10 and STL-10) demonstrates that the proposed method is …