With the development of multiaccess edge computing (MEC) technology, an increasing number of researchers and developers are deploying their computation-intensive and IO-intensive services (especially AI services) on edge devices. These devices, being close to end users, provide better performance in mobile environments. By constructing a service provisioning system at the network edge, latency is significantly reduced due to short-distance communication with edge servers. However, since the MEC-based service provisioning system is resource-sensitive and the network may be unstable, careful resource allocation and traffic scheduling strategies are essential. This paper investigates and quantifies the cost-effectiveness and robustness of the MEC-based service provisioning system with the applied resource allocation and traffic scheduling strategies. Based on this analysis, a cost-effective and robust service provisioning algorithm, termed CERA, is proposed to minimize deployment costs while maintaining system robustness. Extensive experiments are conducted to compare the proposed approach with well-known baseline algorithms and evaluate factors impacting the results. The findings demonstrate that CERA achieves at least 15.9% better performance than other baseline algorithms across various instances.

In the realm of cloud native environments, Ku-bernetes has emerged as the de facto orchestration system for containers, and the service mesh architecture, with its interconnected microservices, has become increasingly prominent. Efficient scheduling and resource allocation for these microservices play a pivotal role in achieving high performance and maintaining system reliability. In this paper, we introduce a novel approach for container scheduling within Kubernetes clusters, leveraging Graph Attention Networks (GATs) for representation learning. Our proposed method captures the intricate dependencies among containers and services by constructing a representation graph. The deep Q-learning algorithm is then employed to optimize scheduling decisions, focusing on container-to-node placements, CPU request-response allocation, and adherence to node affinity and anti-affinity rules. Our experiments demonstrate that our GATs-based method outperforms traditional scheduling strategies, leading to enhanced resource utilization, reduced service latency, and improved overall system throughput. The insights gleaned from this study pave the way for a new frontier in cloud native performance optimization and offer tangible benefits to industries adopting microservice-based architectures.

Military drones can be used for surveillance or spying on enemies. They, however, can be either destroyed or captured, therefore photos contained inside them can be lost or revealed to the attacker. A possible solution to solve such a problem is to adopt Secret Share (SS) techniques to split photos into several sections/chunks and distribute them among a fleet of drones. The advantages of using such a technique are two folds. First, no single drone contains any photo in its entirety; thus even when a drone is captured, the attacker cannot discover any photos. Second, the storage requirements of drones can be simplified, and thus cheaper drones can be produced for such missions. In this scenario, a fleet of drones consists of t+r drones, where t (threshold) is the minimum number of drones required to reconstruct the photos, and r (redundancy) is the maximum number of lost drones the system can tolerate. The optimal configuration of t+r is a formidable task. This configuration is typically rigid and hard to modify in order to fit the requirements of specific missions. In this work, we addressed such an issue and proposed the adoption of a flexible Nested Secret Share (NSS) technique. In our experiments, we compared two of the major SS algorithms (Shamir's schema and the Redundant Residue Number System (RRNS)) with their Two-Level NSS (2NSS) variants to store/retrieve photos. Results showed that Redundant Residue Number System (RRNS) is more suitable for a drone fleet scenario.

Augmented Reality (AR) and Virtual Reality (VR) systems involve computationally intensive image processing algorithms that can burden end-devices with limited resources, leading to poor performance in providing low latency services. Edge-to-cloud computing overcomes the limitations of end-devices by offloading their computations to nearby edge devices or remote cloud servers. Although this proves to be sufficient for many applications, optimal placement of latency sensitive AR/VR services in edge-to-cloud infrastructures (to provide desirable service response times and reliability) remain a formidable challenging. To address this challenge, this paper develops a Multi-Objective Genetic Algorithm (MOGA) to optimize the placement of AR/VR-based services in multi-tier edge-to-cloud environments. The primary objective of the proposed MOGA is to minimize the response time of all running services, while maximizing the reliability of the underlying system from both software and hardware per spectives. To evaluate its performance, we mathematically modeled all components and developed a tailor-made simulator to assess its effectiveness on various scales. MOGA was compared with several heuristics to prove that intuitive solutions, which are usually assumed sufficient, are not efficient enough for the stated problem. The experimental results indicated that MOGA can significantly reduce the response time of deployed services by an average of 67% on different scales, compared to other heuristic methods. MOGA also ensures reliability of the 97% infrastructure (hardware) and 95% services (software).

Emerging computing paradigms like mist, edge, and fog computing address challenges in the real-time processing of vast Internet of Things (IoT) applications. Alongside, cloud computing offers a suitable platform for executing services. Together, they form a multi-tier computing environment known as compute-continuum to efficiently enhance data management and task execution of real-time tasks. The primary considerations for compute-continuum include variations in resource configuration and network architecture, rental cost model, application security needs, energy consumption, transmission latency, and system reliability. To address these problems, we propose two scheduling algorithms (RCSECH and RSECH) for real-time multi-workflow scheduling frameworks. Both algorithms optimize for rental cost, energy consumption, and task reliability when scheduling real-time workflows while considering deadlines and security requirements as constraints. RCSECH also factors in reliability alongside these constraints. The environment under investigation consists of a compute-continuum architecture consisting of mist, edge, fog, and cloud layers, each potentially composed of heterogeneous resources. The framework undergoes evaluation via simulation experiments, revealing promising results. Specifically, the framework exhibits the capability to enhance reliability by up to 7%, reduce energy consumption by 8%, surpass reliability constraints by more than 25%, and generate cost savings by at least 15%.

Industry 4.0 is characterized by digitalized production facilities, where a large volume of sensors collect a vast amount of data that is used to increase the sustainability of the production by e.g. optimizing process parameters, reducing machine downtime and material waste, and the like. However, making intelligent data-driven decisions under timeliness constraints requires the integration of time-sensitive networks with reliable data ingestion and processing infrastructure with plug-in support of Machine Learning (ML) pipelines. However, such integration is difficult due to the lack of frameworks that flexibly integrate and program the networking and computing infrastructures, while allowing ML pipelines to ingest the collected data and make trustworthy decisions in real time. In this paper, we present AIDA - a novel holistic AI-driven network and processing framework for reliable data-driven real-time industrial IoT applications. AIDA manages and configures Time-Sensitive networks (TSN) to enable real-time data ingestion into an observable AI-powered edge/cloud continuum. Pluggable and trustworthy ML components that make timely decisions for various industrial IoT applications and the infrastructure itself are an intrinsic part of AIDA. We introduce the AIDA architecture, demonstrate the building blocks of our framework and illustrate it with two use cases. 

Modern information technology (IT) infrastructures are becoming more complex to meet the diverse demands of emerging technology paradigms such as 5G/6G networks, edge, and internet of things (IoT). The intricacy of these infrastructures grows further when hosting containerized workloads as microservices, resulting in the challenge to detect and troubleshoot performance issues, incidents or even outages of critical use cases like industrial automation processes. Thus, fine-grained measurements and associated visualization are essential for operation observability of these IT infrastructures. However, most existing observability tools operate independently without systematically covering the entire data workflow. This paper presents an integrated design for multi-stage observability workflows, denoted as DistributEd obServability frameworK (DESK). The proposed framework aims to improve observability workflows for measurement, collection, fusion, storage, visualization, and notification. As a proof of concept, we deployed the framework in a Kubernetes-based testbed to demonstrate the successful integration of various components and usability of collected observability data. We also conducted a comprehensive study to determine the caused overhead by DESK agents at the reasonably powerful edge node hardware, which shows on average a CPU and memory overhead of around 2.5 % of total available hardware resource. 

This graduate-level textbook is ideally suited for lecturing the most relevant topics of Edge Computing and its ties to Artificial Intelligence (AI) and Machine Learning (ML) approaches. It starts from basics and gradually advances, step-by-step, to ways AI/ML concepts can help or benefit from Edge Computing platforms. The book is structured into seven chapters; each comes with its own dedicated set of teaching materials (practical skills, demonstration videos, questions, lab assignments, etc.). Chapter 1 opens the book and comprehensively introduces the concept of distributed computing continuum systems that led to the creation of Edge Computing. Chapter 2 motivates the use of container technologies and how they are used to implement programmable edge computing platforms. Chapter 3 introduces ways to employ AI/ML approaches to optimize service lifecycles at the edge. Chapter 4 goes deeper in the use of AI/ML and introduces ways to optimize spreading computational tasks along edge computing platforms. Chapter 5 introduces AI/ML pipelines to efficiently process generated data on the edge. Chapter 6 introduces ways to implement AI/ML systems on the edge and ways to deal with their training and inferencing procedures considering the limited resources available at the edge-nodes. Chapter 7 motivates the creation of a new orchestrator independent object model to descriptive objects (nodes, applications, etc.) and requirements (SLAs) for underlying edge platforms. To provide hands-on experience to students and step-by-step improve their technical capabilities, seven sets of Tutorials-and-Labs (TaLs) are also designed. Codes and Instructions for each TaL is provided on the book website, and accompanied by videos to facilitate their learning process. 

Apache Storm is a distributed processing engine that can reliably process unbounded streams of data for real-time applications. While recent research activities mostly focused on devising a resource allocation and task scheduling algorithm to satisfy high performance or low latency requirements of Storm applications across a distributed and multi-core system, finding a solution that can optimize the energy consumption of running applications remains an important research question to be further explored. In this article, we present a controlling strategy for CPU throttling that continuously optimize the level of consumed energy of a Storm platform by adjusting the voltage and frequency of the CPU cores while running the assigned tasks under latency constraints defined by the end-users. The experimental results running over a Storm cluster with 4 physical nodes (total 24 cores) validates the effectiveness of proposed solution when running multiple compute-intensive operations. In particular, the proposed controller can keep the latency of analytic tasks, in terms of 99th latency percentile, within the quality of service requirement specified by the end-user while reducing the total energy consumption by 18% on average across the entire Storm platform.

To automate network operations and deployment of compute services, intent-driven service management (IDSM) is essential. It enables network users to express their service requirements in a declarative manner as intents. To fulfill the intents, closed control-loop operations carry out required configurations and deployments without human intervention. Despite the fact that intents are fulfilled automatically, conflicts may arise between user's and service provider's intents due to limited resources availability. This triggers IDSM system to initialize an intent negotiation process among conflicting actors. Intent negotiation involves generating one or more alternate intents based on the current state of the underlying physical/virtual resources, which are then presented to the intent creator for acceptance or rejection. In this way, the quality of services (QoS) can be improved significantly by maximizing the acceptance rate of service requests in the scenario of limited resources. However, intent negotiation systems are still in their infancy. The available solutions are platform dependent which poses various challenges in their adoption to diverse platforms. The main focus of this work is to draft and evaluate a comprehensive and generic intent negotiation framework which can be used to develop intent negotiation solutions for diverse IDSM platforms. In this work, we have identified and defined various processes that are necessary for intent negotiation. Furthermore, a generic intent negotiation framework is presented representing interactions among the identified processes, while conflicting actors engage in the intent negotiation. The results demonstrated that the proposed intent negotiation framework increases the intent acceptance rate by up to 38 percent with processing overheads less than 10 percent.