Extended Berkeley Packet Filter (eBPF) is an instruction set and an execution environment inside the Linux kernel. eBPF improves flexibility for data processing and is realized via a virtual machine featuring both a Just-In-Time (JIT) compiler and an interpreter running in the kernel. It executes custom eBPF programs supplied by the user, effectively moving kernel functionality into user space. eBPF has received widespread adoption by companies such as Facebook, Netflix, and academia for a wide range of application domains. eBPF can be used to program the eXpress DataPath (XDP), a kernel network layer that processes packets closer to the NetworkInterface Card (NIC) for fast packet processing. In this thesis, eBPF with XDP, and Iptables, are considered as a Network function(NF), implemented in a Virtual Machine (VM) for packet filtering. The traffic source(source VM) and traffic sink (destination VM) are present in the same subnet. The aim of this thesis is, to understand and investigate the implementation of NFs inVMs and to analyze performance metrics. In VirtualBox, VMs are created to implement the NFs. The results are obtained for the measurements that are essential for the performance evaluation of the NFs, and presented in graphs.

No description available.

No description available.

Discrete Trial Teaching (DTT) is a teaching method that involves fast-paced trials designed to teach basic skills by breaking them into smaller components, typically conducted in a one-on-one setting. Treatment integrity has proven to be of great importance in DTT, with skill acquisition occurring at higher rates when treatment integrity is high. While research has shown that verbal and written feedback are effective in training staff to conduct DTT, there is still a need for research on the use of individualized interventions based on performance assessments. This study used a multiple-probe across participants design, and demonstrated that a one-on-one session including interventions such as feedback, practice, treatment integrity checklists, and/or antecedent interventions is an effective method for increasing treatment integrity and implementation of DTT. The three participants that took part in the individualized interventions all displayed increases in proficiency of delivering DTT trials. / Applied Behavioral Analysis

I examine how managers mitigate the side effects of the overly complicated performance evaluation system in the context of a high-end retail industry. The standard performance evaluation system in the industry has evolved to include multiple performance measures. The detailed measures can incentivize employees to perform multiple performance-relevant activities, but they inevitably increase the complexity of the performance evaluation system. The complexity increases the risk of information overload of the employees, decreasing judgment quality and potentially decreasing their performance. Drawing on psychology literature, I postulate two factors moderating the relationship between information overload and performance: 1) disaggregated feedback provides detailed information on each category of performance measures and compensate for each performance measure rather than for overall performance level; 2) feedforward informs employees about how their actions affect their compensation. Both factors mitigate the negative performance effect of information overload by clarifying causality embedded in the complex performance evaluation system to employees. I conduct two field experiments that implement the disaggregated feedback and the feedforward policies for sales outlets of a high-end retail firm, respectively, and examine whether the policies mitigate information overload problem and improve performance. I find that the treatment group exhibits improvement in performance, suggesting that disaggregated feedback and the feedforward reduce information overload. / Business Administration/Accounting

High-quality drinking water and wastewater systems are essential to public health, business, and quality of life in the United States. Even though the current performance of these systems is moderate, the concern is about the future performance. Planning can be done for improvement once the current performance of utilities is evaluated, and areas with a scope of improvement are identified. Benchmarking and performance evaluation are key components in the process of continuous improvement for utility's performance. Benchmarking helps utilities make policies and programmatic decisions that reduce operational expenses and increase productivity by understanding areas of underperformance, understanding customer needs, developing future plans, and setting goals. This study establishes a strong case for implementing benchmarking methodologies among utilities to evaluate and improve performance. There are many initiatives on performance benchmarking of utilities but a few of them focuses on one or few area of performance. There are a few initiatives which use subjective indicators. Additionally, consultants visit the utilities for performance evaluation. This research focuses on creating a web-based benchmarking platform for performance evaluation using holistic and quantitative indicators. Practical and robust methodologies are used and the research presents the current performance comparisons among utilities for areas that impact overall utility's performance. Web based benchmarking consists of two major parts -- data collection and result visualization. A major contribution from this study is the creation of an online performance benchmarking database. With time more data will be collected which will provide utilities an access to a better database for performance evaluation. The future work in this research will be analyzing the data and results for each participant for each set of indicators, and finding possible reasons for under performance, followed by suggesting solutions for improvement using the best practices. / Master of Science

The emergence of scientific applications embedded with multiple modes of parallelism has made heterogeneous computing systems indispensable in high performance computing. The popularity of such systems is evident from the fact that three out of the top five fastest supercomputers in the world employ heterogeneous computing, i.e., they use dissimilar computational units. A closer look at the performance of these supercomputers reveals that they achieve only around 50% of their theoretical peak performance. This suggests that applications that were tuned for erstwhile homogeneous computing may not be efficient for today's heterogeneous computing and hence, novel optimization strategies are required to be exercised. However, optimizing an application for heterogeneous computing systems is extremely challenging, primarily due to the architectural differences in computational units in such systems. This thesis intends to act as a cookbook for optimizing applications on heterogeneous computing systems that employ graphics processing units (GPUs) as the preferred mode of accelerators. We discuss optimization strategies for multicore CPUs as well as for the two popular GPU platforms, i.e., GPUs from AMD and NVIDIA. Optimization strategies for NVIDIA GPUs have been well studied but when applied on AMD GPUs, they fail to measurably improve performance because of the differences in underlying architecture. To the best of our knowledge, this research is the first to propose optimization strategies for AMD GPUs. Even on NVIDIA GPUs, there exists a lesser known but an extremely severe performance pitfall called partition camping, which can affect application performance by up to seven-fold. To facilitate the detection of this phenomenon, we have developed a performance prediction model that analyzes and characterizes the effect of partition camping in GPU applications. We have used a large-scale, molecular modeling application to validate and verify all the optimization strategies. Our results illustrate that if appropriately optimized, AMD and NVIDIA GPUs can provide 371-fold and 328-fold improvement, respectively, over a hand-tuned, SSE-optimized serial implementation. / Master of Science

In order to operate UAVs under human supervisory control in more complex arenas such as urban environments, an obstacle detection system is a requirement to achieve safe navigation. The development of a system capable of meeting these requirements is presented. The first stage of development was sensor selection and initial testing. After this, the sensor was combined with a servomotor to allow it to rotate and provide obstacle detection coverage in front, below, and to both sides of the UAV. Utilizing a PC-104 single board computer running LabView Real-time for on-board control of the sensor and servomotor, a stand alone obstacle detection system was developed meeting the requirements of light weight, low power, and small size. The detection performance of the system for several parameters has been fully characterized. A human subjects study was conducted to assess the any advantages resulting from the addition of the obstacle detection system compared to that of a normal nadir camera. The study demonstrated that users with access to the three-dimensional display were able to navigate an obstacle course with greater success than those with only a camera. Additional development into more advanced visualization of the environment has potential to increase effectiveness of this obstacle detection system. / Master of Science

This paper inspired by simulation by Artificial Neural Networks (ANNs) applied recently for evaluation of phonics methodology to teach children "how to read". A novel approach for teaching a mathematical topic using a computer aided learning (CAL) package applied at educational field (a children classroom). Interesting practical results obtained after field application of suggested CAL package with and without associated teacher's voice. Presented study highly recommends application of a novel teaching trend based on behaviorism and individuals' learning styles. That is to improve quality of children mathematical learning performance.

The thesis Performance Analysis of employees on the Service Desk to be engaged in evaluating and rewarding performance of employees working on the Service Desk. The work is divided into two parts. The theoretical part examines possible ways of evaluating and rewarding performance of employees. The practical part proposes a solution which demonstrates the real performance of employees, set of financial rewards commensurate with their performance and propose the implementation of modified assessment and reward performance.