Search

Typical mobile edge computing infrastructures have to contend with unreliable computing devices at their end-points. The limited resource capacities of mobile edge devices gives rise to frequent contentions, node overloads or failures. This is exacerbated by the strict deadlines of modern applications. To avoid failures, fault-tolerant approaches utilize preemptive migration to transfer active tasks across nodes and prevent nodes running at capacity. However, prior work struggles to dynamically adapt in settings with highly volatile workloads or even accurately detect and diagnose anomalies for optimal remediation. To meet the strict service level objectives of contemporary workloads, there is a need for dynamic fault-tolerant methods that can quickly adapt to changes in edge environments while having parsimonious remediation in the form of preemptive migration to avoid stressing the system network. This work proposes PreGAN, featuring a Generative Adversarial Network (GAN) based approach to predict contentions, pinpoint specific resource types with high chance of overload, and generate migration decisions to proactively avoid system downtime. PreGAN leverages coupled-simulations to train the GAN model at run-time and a few-shot fault classifier to update decisions of an underpinning scheduler. We also extend it to PreGAN+ that also periodically tunes the decision model using semi-supervised training and a Transformer based neural network for low tuning time, albeit with higher memory overheads. Experiments on a Raspberry-Pi based edge environment demonstrate that both models outperform state-of-the-art baselines in fault detection and diagnosis scores by up to 12.5% and 31.2% respectively. This also translates in improvements in Quality of Service against baseline approaches.

The rise of distributed cloud computing technologies has been pivotal for the large-scale adoption of Artificial Intelligence (AI) based applications for high fidelity and scalable service delivery. Systematic resource management is central in maintaining optimal Quality of Service (QoS) in cloud platforms and is divided into three fundamental types: resource provisioning, AI model deployment and workload placement. To exploit the synergy among these decision types, it becomes imperative to concurrently design (co-design) the provisioning, deployment and placement decisions for optimal QoS. As users and cloud service providers shift to non-stationary AI-based workloads, frequent decision making imposes severe time constraints on the resource management models. Existing AI-based solutions often optimize decision types independently and tend to ignore the dependencies across various system performance aspects such as energy consumption and CPU utilization, making them perform poorly in large-scale cloud systems. To address this, we propose a novel method, called SciNet, that leverages a co-simulated digital-twin of the infrastructure to capture inter-metric dependencies and accurately estimate QoS scores. To avoid expensive simulation overheads at test time, SciNet trains a neural network based imitation learner that aims to mimic an oracle, which takes optimal decisions based on co-simulated QoS estimates. Offline model training and online decision making based on the imitation learner, enables SciNet to take optimal decisions while being time-efficient. Experiments with real-life AI-based benchmark applications on a public cloud testbed show that SciNet gives up to 48% lower execution cost, 79% higher inference accuracy, 71% lower energy consumption and 56% lower response times compared to the current state-of-the-art methods.

In recent years, deep learning models have become ubiquitous in industry and academia alike. Deep neural networks can solve some of the most complex pattern-recognition problems today, but come with the price of massive compute and memory requirements. This makes the problem of deploying such large-scale neural networks challenging in resource-constrained mobile edge computing platforms, specifically in mission-critical domains like surveillance and healthcare. To solve this, a promising solution is to split resource-hungry neural networks into lightweight disjoint smaller components for pipelined distributed processing. At present, there are two main approaches to do this: semantic and layer-wise splitting. The former partitions a neural network into parallel disjoint models that produce a part of the result, whereas the latter partitions into sequential models that produce intermediate results. However, there is no intelligent algorithm that decides which splitting strategy to use and places such modular splits to edge nodes for optimal performance. To combat this, this work proposes a novel AI-driven online policy, SplitPlace, that uses Multi-Armed-Bandits to intelligently decide between layer and semantic splitting strategies based on the input task's service deadline demands. SplitPlace places such neural network split fragments on mobile edge devices using decision-aware reinforcement learning for efficient and scalable computing. Moreover, SplitPlace fine-tunes its placement engine to adapt to volatile environments. Our experiments on physical mobile-edge environments with real-world workloads show that SplitPlace can significantly improve the state-of-the-art in terms of average response time, deadline violation rate, inference accuracy, and total reward by up to 46, 69, 3 and 12 percent respectively.

Cloud computing is a mainstay of modern technology, offering cost-effective and scalable solutions to a variety of different problems. The massive shift of organization resource needs from local systems to cloud-based systems has greatly increased the costs incurred by cloud providers in expanding, maintaining, and supplying server, storage, network, and processing hardware. Due to the large scale at which cloud providers operate, even small performance degradation issues can cause energy or resource usage costs to rise dramatically. One way in which cloud providers may improve cost reduction is by reducing energy consumption. The use of intelligent task-scheduling algorithms to allocate user-deployed jobs to servers can reduce the amount of energy consumed. Conventional task scheduling algorithms involve both heuristic and metaheuristic methods. Recently, the application of Artificial Intelligence (AI) to optimize task scheduling has seen significant progress, including the Gated Graph Convolution Network (GGCN). This paper proposes a new approach called HunterPluswhich examine the effect of extending the GGCN’s Gated Recurrent Unit to a Bidirectional Gated Recurrent Unit. The paper also studies the utilization of Convolutional Neural Networks (CNNs) in optimizing cloud–fog task scheduling. Experimental results show that the CNN scheduler outperforms the GGCN-based models in both energy consumption per task and job completion rate metrics by at least 17 and 10.4 percent, respectively.

•Propose a novel scheme to predict the impact of COVID-19 Pandemic.•Design a model based on Cloud Computing and Machine Learning for real-time prediction.•Show improved prediction accuracy compared to baseline method.•Highlight key future research directions and emerging trends.

•Transformative effects of emerging paradigms and technologies on cloud computing systems.•To study the history and evolution of cloud computing from year 1958.•To identify the influence of triumvirate “IoT + AI + Blockchain” on cloud computing.•Vision and trends for academicians, practitioners and researchers working in cloud.•Cloud futurology model to manage next-generation computing systems holistically.