The novelty of this dissertation is the optimization and acceleration of deep neural networks aimed at real-time predictions with minimal energy consumption. It consists of cross-layer optimization, output directed dynamic quantization, and opportunistic near-data computation for deep neural network acceleration. On two datasets (CIFAR-10 and CIFAR-100), the proposed deep neural network optimization and acceleration frameworks are tested using a variety of Convolutional neural networks (e.g., LeNet-5, VGG-16, GoogLeNet, DenseNet, ResNet). Experimental results are promising when compared to other state-of-the-art deep neural network acceleration efforts in the literature.

Forty percent of employed parents report that they experience work-family conflict (Galinsky, Bond, & Friedman, 1993). Work-family conflict (WFC) exists when role pressures from the work and family domains are mutually incompatible. WFC is associated with decreases in family, job, and life satisfaction and physical health; intention to quit one's job; and increases in workplace absenteeism. Women may be more impacted by WFC than men, as women report completing 65-80% of the child care (Sayer, 2001) and spend 80 hours per week fulfilling work and home responsibilities (Cowan, 1983). Research suggests that WFC can be reduced with social support, such as co-workers providing assistance when family interferes with work (Carlson & Perrewe, 1999). It is unclear whether parents 'get a break' or are penalized by co-workers. The purpose of the present study was to examine co-workers' reactions to individuals who experience WFC. Based on sex role theory and attribution theory, it was predicted that women, people who experience family interference with work, and those who have more control over the work interference would be helped less and evaluated more poorly on a team task than men, people who experience non-family related work interference, and those who have less control over the work interference. A laboratory experiment was conducted in which participants signed up for a team-based study. The teammate was a confederate who was late for the study. Teammate control over the tardiness (unexpected physician's visit versus forgotten physician's appointment), type of work conflict (self- versus family-related), and gender of the teammate were manipulated. After learning about the reasons for the tardiness of their teammate, the 218 participants (63% female; 59% Caucasian) decided whether to help the late teammate by completing a word sort task for them or letting the late teammate make up the work after the experiment. When the teammate arrived, the participants completed a team task and then evaluated the task performance of their teammate. None of the hypotheses were confirmed in this study. However, exploratory analyses showed that people who had more control over the tardiness were rated lower than people who had less control over the tardiness. Contrary to expectations, exploratory analyses also showed that men rated women who were late to the study for a family-related reason higher than women who were late due to a self-related reason. These findings suggest that male co-workers may give women a break when they experience family interference with work. Implications for future research and practice are discussed.

Peer-to-peer swarming, as used by BitTorrent, is one of the de facto solutions for content dissemination in today’s Internet. By leveraging resources provided by users, peer-to-peer swarming is a simple, scalable and efficient mechanism for content distribution. Although peer-to-peer swarming has been widely studied for a decade, prior work has focused on the dissemination of one commodity (a single file). This thesis focuses on the multi-commodity case. We have discovered through measurements that a vast number of publishers currently disseminate multiple files in a single swarm (bundle). The first contribution of this thesis is a model for content availability. We use the model to show that, when publishers are intermittent, bundling K files increases content availability exponentially as function of K. When there is a stable publisher, we consider content availability among peers (excluding the publisher). Our second contribution is the estimate of the dependency of peers on the stable publisher, which is useful for provisioning purposes as well as in deciding how to bundle. To this goal, we propose a new metric, swarm self-sustainability, and present a model that yields swarm self-sustainability as a function of the file size, popularity and service capacity of peers. Then, we investigate reciprocity and the use of barter that occurs among peers. As our third contribution, we prove that the loss of efficiency due to the download of unrequested content to enforce direct reciprocity, as opposed to indirect reciprocity, is at most two in a class of networks without relays. Finally, we study algorithmic and economic problems faced by enterprises who leverage swarming systems and who control prices and bundling strategies. As our fourth contribution, we present two formulations of the optimal bundling problem, and prove that one is NP hard whereas the other is solvable by a greedy strategy. From an economic standpoint, we present conditions for the existence and uniqueness of an equilibrium between publishers and peers.

QUIC is a new secure multiplexed transport protocol built on top of UDP. This general-purpose transport protocol aims to provide the lowest connection latency possible and solve the shortcomings of TCP, UDP, and current problems of the internet. Furthermore, it allows further development of the transport protocol without upgrading the network infrastructure. Last year in May 2021, QUIC was finally standardized by the IETF, allowing for full development and release while also opening the path for future research as older research dated due to using the older version and the finalization of QUIC standard protocol. While there are a lot of different QUIC implementations, this thesis selected two and conducted a performance evaluation on the cloud environment and compared the two while also taking the sustainability aspect into account. Asa result, we will find which of the selected implementation is environmentally friendly through this experiment while also providing good performance. / <p>2022 GENIAL Summer School</p>

This paper presents a Call Admission Control (CAC) algorithm based fuzzy logic to maintain the quality of service using cell breathing concept. When a new call is accepted by a cell, its current user is generally affected due to cell breathing. The proposed CAC algorithm accepts a new call only if the current users in the cell are not jeopardized. Performance evaluation is done for single-cell and multicell scenarios. In multicell scenario dynamic assignment of users to the neighboring cell, so called handoff, has been considered to achieve a lower blocking probability. Handoff and new call requests are assumed with handoff being given preference using a reserved channel scheme. CAC for different types of services are shown which depend upon the bandwidth requirement for voice, data and video. Distance, arrival rate, bandwidth and nonorthogonality factor of the signal are considered for making the call acceptance decision. The paper demonstrates that fuzzy logic with the cell breathing concept can be used to develop a CAC algorithm to achieve a better performance evaluation.

Quality of Service (QoS) routing considered as one of the major components of the QoS framework in communication networks. The concept of QoS routing has emerged from the fact that routers direct traffic from source to destination, depending on data types, network constraints and requirements to achieve network performance efficiency. It has been introduced to administer, monitor and improve the performance of computer networks. Many QoS routing algorithms are used to maximize network performance by balancing traffic distributed over multiple paths. Its major components include bandwidth, delay, jitter, cost, and loss probability in order to measure the end users¿ requirements, optimize network resource usage and balance traffic load. The majority of existing QoS algorithms require the maintenance of the global network state information and use it to make routing decisions. The global QoS network state needs to be exchanged periodically among routers since the efficiency of a routing algorithm depends on the accuracy of link-state information. However, most of QoS routing algorithms suffer from scalability problems, because of the high communication overhead and the high computation effort associated with marinating and distributing the global state information to each node in the network.The goal of this thesis is to contribute to enhancing the scalability of QoS routing algorithms. Motivated by this, the thesis is focused on localized QoS routing that is proposed to achieve QoS guarantees and overcome the problems of using global network state information such as high communication overhead caused by frequent state information updates, inaccuracy of link-state information for large QoS state update intervals and the route oscillating due to the view of state information. Using such an approach, the source node makes its own routing decisions based on the information that is local to each node in the path. Localized QoS routing does not need the global network state to be exchanged among network nodes because it infers the network state and avoids all the problems associated with it, like high communication and processing overheads and oscillating behaviour. In localized QoS routing each source node is required to first determine a set of candidate paths to each possible destination. In this thesis we have developed localized QoS routing algorithms that select a path based on its quality to satisfy the connection requirements. In the first part of the thesis a localized routing algorithm has been developed that relies on the average residual bandwidth that each path can support to make routing decisions. In the second part of the thesis, we have developed a localized delay-based QoS routing (DBR) algorithm which relies on a delay constraint that each path satisfies to make routing decisions. We also modify credit-based routing (CBR) so that this uses delay instead of bandwidth. Finally, we have developed a localized QoS routing algorithm for routing in two levels of a hierarchal network and this relies on residual bandwidth to make routing decisions in a hierarchical network like the internet. We have compared the performance of the proposed localized routing algorithms with other localized and global QoS routing algorithms under different ranges of workloads, system parameters and network topologies. Simulation results have indicated that the proposed algorithms indeed outperform algorithms that use the basics of schemes that currently operate on the internet, even for a small update interval of link state. The proposed algorithms have also reduced the routing overhead significantly and utilize network resources efficiently.

The Quality of Service (QoS) is a profound concept which is gaining increasing attention in the Internet industry. Best-effort applications are now no longer acceptable in certain situations needing high bandwidth provisioning, low loss and streaming of multimedia applications. New emerging multimedia applications are requiring new levels of quality of services beyond those supported by best-effort networks. Quality of service routing is an essential part in any QoS architecture in communication networks. QoS routing aims to select a path among the many possible choices that has sufficient resources to accommodate the QoS requirements. QoS routing can significantly improve the network performance due to its awareness of the network QoS state. Most QoS routing algorithms require maintenance of the global network¿s state information to make routing decisions. Global state information needs to be periodically exchanged among routers since the efficiency of a routing algorithm depends on link-state information accuracy. However, most QoS routing algorithms suffer from scalability due to the high communication overhead and the high computation effort associated with maintaining accurate link state information and distributing global state information to each node in the network. The ultimate goal of this thesis is to contribute towards enhancing the scalability of QoS routing algorithms. Towards this goal, the thesis is focused on Localised QoS routing algorithms proposed to overcome the problems of using global network state information. Using such an approach, the source node makes routing decisions based on the local state information for each node in the path. Localised QoS routing algorithms avoid the problems associated in the global network state, like high communication and processing overheads. In Localised QoS routing algorithms each source node maintains a predetermined set of candidate paths for each destination and avoids the problems associated with the maintenance of a global network state by using locally collected flow statistics and flow blocking probabilities. / Libya's higher education

The McMaster incident detection algorithm is being tested on-line within the Burlington freeway traffic management system (FTMS) as an alternative to the existing California-type algorithm currently in place. This paper represents the most recent and comprehensive evaluation of the McMaster algorithm's performance to date. In the past, the algorithm has been tested using single lane detectors for the northbound lanes only. This evaluation uses data from lanes 1 and 2 for each of the 13 northbound and 13 southbound detector stations. The data was collected during a 60-day period beginning on November 15, 1990 and ending January 13, 1991. Detection rate, mean time-lag to detection and false alarm rate are used to evaluate the performance of the algorithm. As well, those factors such as winter precipitation, which influenced the performance of the algorithm are also examined. To improve the algorithm's detection rate and lower its false alarm rate, it is reccomended that the persistence check used to declare an incident be increased by 30-seconds from 2 to 3 periods. / Thesis / Candidate in Philosophy

<p dir="ltr">Practically, Byzantine faults are not assumed in cloud applications. Byzantine fault-tolerance adds significant cryptographic, communication, throughput, and latency overheads to applications, contributing to the resistance towards its widespread adoption. Existing Byzantine-fault tolerant protocols focus on optimal latency or optimal communication while ignoring the throughput and cryptographic overheads.</p><p dir="ltr">In this thesis, we explore pipelining for Byzantine fault-tolerant applications. Pipelining tasks is a common optimization in distributed systems that involves executing tasks in stages. The idea is that instead of executing a task in an iteration as an atomic unit, we split the execution into stages and execute all stages of <i>different</i> tasks per iteration. We observe significant performance benefits if executing later stages of a task helps other tasks in earlier stages, saving effort in each stage. The length of the pipeline, i.e., the number of stages, determines the latency of an individual task. However, if the pipeline improves the execution of every stage enough, then the latency improves.</p><p dir="ltr">We primarily explore three Byzantine Fault Tolerant (BFT) applications with pipelining: (i) unique chain-based State Machine Replication protocols: <i>Apollo</i>, <i>Artemis</i>, <i>Leto</i>, and <i>Zeus</i>, and (ii) energy-efficient State Machine Replication: <i>EESMR</i>. (iii) random beacon protocols: <i>GRandPiper</i>, <i>BRandPiper</i>, and <i>OptRand</i>. We design them with a pipeline-first approach to improve the throughput, cryptographic, and communication costs at every stage of the pipeline. With respect to latency, we show (i) pipelined SMR protocols where our pipeline stages have constant cryptographic and linear communication costs allowing our protocols to outperform state-of-the-art BFT-SMR protocols in throughput. (ii) pipelined SMR protocols with techniques to make each stage of the pipeline independent, thus achieving demonstrable energy efficiency while allowing an unbounded number of non-interactive parallel proposals. (iii) reduced latencies for reconfiguration-friendly random beacons by using two pipelines: an SMR pipeline to commit and a beacon pipeline to produce random numbers and decoupling the two pipelines thereby removing the impact of the high-latency SMR pipeline on the latency of the randomness output by the system. </p>

Robots are often used within the industry for automated tasks that are too dangerous, complex, or strenuous for humans, which leads to time and cost benefits. Robots can have an arm and a gripper to manipulate the world and sensors for eyes to be able to perceive the world. Human vision can be seen as an effortless task, but machine vision requires substantial computation in an attempt to be as effective as human vision. Visual object recognition is a common goal for machine vision, and it is often applied using deep learning and generic object detection. This thesis has a focus on robot guidance systems that include a robot with its gripper on the robot arm, a camera that acquires images of the world, boxes to detect in one or more layers, and the software that applies a generic object detection model to detect the boxes. Robot guidance systems’ performance is impacted by many variables such as different environmental, camera, object, and robot gripper aspects. A survey was constructed to receive feedback from professionals on what thresholds that can be defined for detection from the model to be counted as correct, with the aspect of the detection referring to an actual object that needs to be able to be picked up by a robot. This thesis has implemented precision, recall, average precision at a specific threshold, average precision at a range of thresholds, localization-recall-precision error, and a manually constructed counter based on survey results for the robot’s ability to pick up an object from the information provided by the detection, called pickability score. The metrics from this thesis are implemented within a tool intended for analyzing different models’ performance on varying datasets. The values of all the metrics for the applied dataset are presented in the results. The metrics are discussed with regards to what information they portray together with a robot guidance system. The conclusion is to see the metrics for what they are best at by themselves. Use the average precision metrics for the performance evaluation of the models, and the pickability scores with extended features for the robot gripper pickability evaluation.