The rapid growth of data generated from Internet of Things (IoTs) such as smart phones and smart home devices presents new challenges to cloud computing in transferring, storing, and processing the data. With increasingly more powerful edge devices, edge computing, on the other hand, has the potential to better responsiveness, privacy, and cost efficiency. However, resources across the cloud and edge are highly distributed and highly diverse. To address these challenges, this paper proposes EdgeFaaS, a Function-as-a-Service (FaaS) based computing framework that supports the flexible, convenient, and optimized use of distributed and heterogeneous resources across IoT, edge, and cloud systems. EdgeFaaS allows cluster resources and individual devices to be managed under the same framework and provide computational and storage resources for functions. It provides virtual function and virtual storage interfaces for consistent function management and storage management across heterogeneous compute and storage resources. It automatically optimizes the scheduling of functions and placement of data according to their performance and privacy requirements. EdgeFaaS is evaluated based on two edge workflows: video analytics workflow and federated learning workflow, both of which are representative edge applications and involve large amounts of input data generated from edge devices.

Personalized learning is gaining popularity in online computer science education due to its characteristics of pacing the learning progress and adapting the instructional approach to each individual learner from a diverse background. Among various instructional methods in computer science education, hands-on labs have unique requirements of understanding learners' behavior and assessing learners' performance for personalization. Hands-on labs are a critical learning approach for cybersecurity education. It provides real-world complex problem scenarios and helps learners develop a deeper understanding of knowledge and concepts while solving real-world problems. But there are unique challenges when using hands-on labs for cybersecurity education. Existing hands-on lab exercises materials are usually managed in a problem-centric fashion, while it lacks a coherent way to manage existing labs and provide productive lab exercising plans for cybersecurity learners. To solve these challenges, a personalized learning platform called ThoTh Lab specifically designed for computer science hands-on labs in a cloud environment is established. ThoTh Lab can identify the learning style from student activities and adapt learning material accordingly. With the awareness of student learning styles, instructors are able to use techniques more suitable for the specific student, and hence, improve the speed and quality of the learning process. ThoTh Lab also provides student performance prediction, which allows the instructors to change the learning progress and take other measurements to help the students timely. A knowledge graph in the cybersecurity domain is also constructed using Natural language processing (NLP) technologies including word embedding and hyperlink-based concept mining. This knowledge graph is then utilized during the regular learning process to build a personalized lab recommendation system by suggesting relevant labs based on students' past learning history to maximize their learning outcomes. To evaluate ThoTh Lab, several in-class experiments were carried out in cybersecurity classes for both graduate and undergraduate students at Arizona State University and data was collected over several semesters. The case studies show that, by leveraging the personalized lab platform, students tend to be more absorbed in a lab project, show more interest in the cybersecurity area, spend more effort on the project and gain enhanced learning outcomes.

Demand for processing machine learning workloads has grown incredibly over the past few years. Kubernetes, an open-source container orchestrator, has been widely used by public and private cloud providers for building scalable systems for meeting this demand. The data used to train machine learning workloads can be sensitive in nature, and organizations may prefer to be responsible for their data security and governance by housing it on on-premises systems. Hybrid cloud gives organizations the flexibility to use both on-premises and cloud infrastructure together, leveraging the advantages of both. While there is a long list of benefits, Kubernetes has limitations by design that limit a user’s abilities in a hybrid cloud environment. The Kubernetes control plane does not allow for the management of worker nodes across cloud providers. This boundary puts new responsibilities on the end-user when deploying a hybrid cloud workload. The end-user must create their clusters and specify which cluster the workload will be scheduled to ahead of time. The Kubernetes scheduler will not take the capacity of another cluster into account. To address these limitations, this thesis presents a new hybrid cloud Kubernetes scheduler that can create new clusters on-demand and burst machine learning workloads to a public cloud when on-premises resources are insufficient. Workloads begin scheduling on an on-premises Kubernetes cluster. When the on-premises cluster’s capacity is exhausted, a new Kubernetes cluster is created on-demand in a public cloud provider, and machine learning tasks waiting in the Kubernetes scheduling queue are dynamically migrated to the public cloud provider’s Kubernetes cluster. The public Kubernetes cluster is dynamically sized and auto scaled based on the pending tasks’ demand. When migrating tasks, the data dependencies among tasks are considered, and a region is dynamically chosen to reduce migration time and cost. The scheduler is experimentally evaluated with real-world machine learning workloads, including predicting if a subscriber will stay with a subscription service, predicting the discount needed to retain a subscription customer, predicting if a credit card transaction is fraudulent, and simulated real-world job arrival behavior in a real hybrid cloud environment. Results show that the scheduler can substantially reduce the workload execution time by dynamically migrating tasks from on-premises to public cloud and minimizing the cost by dynamically sizing and scaling the public cluster.

The rapid growth of Internet-of-things (IoT) and artificial intelligence applications have called forth a new computing paradigm--edge computing. Edge computing applications, such as video surveillance, autonomous driving, and augmented reality, are highly computationally intensive and require real-time processing. Current edge systems are typically based on commodity general-purpose hardware such as Central Processing Units (CPUs) and Graphical Processing Units (GPUs) , which are mainly designed for large, non-time-sensitive jobs in the cloud and do not match the needs of the edge workloads. Also, these systems are usually power hungry and are not suitable for resource-constrained edge deployments. Such application-hardware mismatch calls forth a new computing backbone to support the high-bandwidth, low-latency, and energy-efficient requirements. Also, the new system should be able to support a variety of edge applications with different characteristics. This thesis addresses the above challenges by studying the use of Field Programmable Gate Array (FPGA) -based computing systems for accelerating the edge workloads, from three critical angles. First, it investigates the feasibility of FPGAs for edge computing, in comparison to conventional CPUs and GPUs. Second, it studies the acceleration of common algorithmic characteristics, identified as loop patterns, using FPGAs, and develops a benchmark tool for analyzing the performance of these patterns on different accelerators. Third, it designs a new edge computing platform using multiple clustered FPGAs to provide high-bandwidth and low-latency acceleration of convolutional neural networks (CNNs) widely used in edge applications. Finally, it studies the acceleration of the emerging neural networks, randomly-wired neural networks, on the multi-FPGA platform. The experimental results from this work show that the new generation of workloads requires rethinking the current edge-computing architecture. First, through the acceleration of common loops, it demonstrates that FPGAs can outperform GPUs in specific loops types up to 14 times. Second, it shows the linear scalability of multi-FPGA platforms in accelerating neural networks. Third, it demonstrates the superiority of the new scheduler to optimally place randomly-wired neural networks on multi-FPGA platforms with 81.1 times better throughput than the available scheduling mechanisms.

Data-driven simulations represent a promising approach to understanding and predicting complex dynamic processes in the presence of shifting demands of urban systems. Newly proposed continuous ensemble frameworks like DataStorm, execute the models in coupled continuous simulations, thus producing multiple outputs per execution of the model, i.e., any time instant may be covered by multiple simulation ensembles of the corresponding model, each with a different set of data and parameter values. Continuous frameworks focus on designing ensemble configurations that appropriately and efficiently cover the input parameter space, and do not give importance to building causal narratives during continuous execution, which is essentially important for understanding and interpreting the end simulation data. The thesis aims to address this challenge by building causal-fabrics during continuous execution, which essentially contributes to causal-creation of simulation ensembles, which are similar to its previous history, input sample-parameter and output-stream explored. I introduce the following metrics: Provenance-similarity, Output-similarity, and parametric-similarity which could be used to weave such acausal-fabric into explainable causal-narratives. The DataStorm execution framework runs the models in different ensemble configurations to cover the input parameter space efficiently. An end-user may have a preference for a parameter subspace and may seek to find causal narratives where the ensembles in the causal narratives reside in that parameter sub-space. I present an additional constraint called Preference Query E, that defines such a preferred input parameter subspace. I propose a new method of ensemble creation, where during each execution step in continuous execution, more bias is given to the creation of new ensembles that are present in that preferred parameter subspace. Once such narratives have been constructed, I use the concept of timelines to build causal explorations of the causal narratives that have been constructed. I present an approximate top-K timelines algorithm that discovers K such timelines using heuristic search. The next step is to find time-lines that are causal-explorations in the preferred parameter subspace. Using a probabilistic skyline query, I aim to discover a subset of top-K timelines where each timeline has the maximum number of simulation ensembles present in the desired parameter subspace. These subsets of timelines effectively describe such causal exploration in the preferred parameter subspace.

The meteoric rise of Deep Neural Networks (DNN) has led to the development of various Machine Learning (ML) frameworks (e.g., Tensorflow, PyTorch). Every ML framework has a different way of handling DNN models, data types, operations involved, and the internal representations stored on disk or memory. There have been initiatives such as the Open Neural Network Exchange (ONNX) for a more standardized approach to machine learning for better interoperability between the various popular ML frameworks. Model Serving Platforms (MSP) (e.g., Tensorflow Serving, Clipper) are used for serving DNN models to applications and edge devices. These platforms have gained widespread use for their flexibility in serving DNN models created by various ML frameworks. They also have additional capabilities such as caching, automatic ensembling, and scheduling. However, few of these frameworks focus on optimizing the storage of these DNN models, some of which may take up to ∼130GB storage space(“Turing-NLG: A 17-billion-parameter language model by Microsoft” 2020). These MSPs leave it to the ML frameworks for optimizing the DNN model with various model compression techniques, such as quantization and pruning. This thesis investigates the viability of automatic cross-model compression using traditional deduplication techniques and storage optimizations. Scenarios are identified where different DNN models have shareable model weight parameters. “Chunking” a model into smaller pieces is explored as an approach for deduplication. This thesis also proposes a design for storage in a Relational Database Management System (RDBMS) that allows for automatic cross-model deduplication.

Infectious diseases spread at a rapid rate, due to the increasing mobility of the human population. It is important to have a variety of containment and assessment strategies to prevent and limit their spread. In the on-going COVID-19 pandemic, telehealth services including daily health surveys are used to study the prevalence and severity of the disease. Daily health surveys can also help to study the progression and fluctuation of symptoms as recalling, tracking, and explaining symptoms to doctors can often be challenging for patients. Data aggregates collected from the daily health surveys can be used to identify the surge of a disease in a community. This thesis enhances a well-known boosting algorithm, XGBoost, to predict COVID-19 from the anonymized self-reported survey responses provided by Carnegie Mellon University (CMU) - Delphi research group in collaboration with Facebook. Despite the tremendous COVID-19 surge in the United States, this survey dataset is highly imbalanced with 84% negative COVID-19 cases and 16% positive cases. It is tedious to learn from an imbalanced dataset, especially when the dataset could also be noisy, as seen commonly in self-reported surveys. This thesis addresses these challenges by enhancing XGBoost with a tunable loss function, ?-loss, that interpolates between the exponential loss (? = 1/2), the log-loss (? = 1), and the 0-1 loss (? = ∞). Results show that tuning XGBoost with ?-loss can enhance performance over the standard XGBoost with log-loss (? = 1).

Contact tracing has been shown to be effective in limiting the rate of spread of infectious diseases like COVID-19. Several solutions based on the exchange of random, anonymous tokens between users’ mobile devices via Bluetooth, or using users’ location traces have been proposed and deployed. These solutions require the user device to download the tokens (or traces) of infected users from the server. The user tokens are matched with infected users’ tokens to determine an exposure event. These solutions are vulnerable to a range of security and privacy issues, and require large downloads, thus warranting the need for an efficient protocol with strong privacy guarantees. Moreover, these solutions are based solely on proximity between user devices, while COVID-19 can spread from common surfaces as well. Knowledge of areas with a large number of visits by infected users (hotspots) can help inform users to avoid those areas and thereby reduce surface transmission. This thesis proposes a strong secure system for contact tracing and hotspots histogram computation. The contact tracing protocol uses a combination of Bluetooth Low Energy and Global Positioning System (GPS) location data. A novel and deployment-friendly Delegated Private Set Intersection Cardinality protocol is proposed for efficient and secure server aided matching of tokens. Secure aggregation techniques are used to allow the server to learn areas of high risk from location traces of diagnosed users, without revealing any individual user’s location history.

Apache Spark is one of the most widely adopted open-source Big Data processing engines. High performance and ease of use for a wide class of users are some of the primary reasons for the wide adoption. Although data partitioning increases the performance of the analytics workload, its application to Apache Spark is very limited due to layered data abstractions. Once data is written to a stable storage system like Hadoop Distributed File System (HDFS), the data locality information is lost, and while reading the data back into Spark’s in-memory layer, the reading process is random which incurs shuffle overhead. This report investigates the use of metadata information that is stored along with the data itself for reducing shuffle overload in the join-based workloads. It explores the Hyperspace library to mitigate the shuffle overhead for Spark SQL applications. The article also introduces the Lachesis system to solve the shuffle overhead problem. The benchmark results show that the persistent partition and co-location techniques can be beneficial for matrix multiplication using SQL (Structured Query Language) operator along with the TPC-H analytical queries benchmark. The study concludes with a discussion about the trade-offs of using integrated stable storage to layered storage abstractions. It also discusses the feasibility of integration of the Machine Learning (ML) inference phase with the SQL operators along with cross-engine compatibility for employing data locality information.