Vehicular Edge Computing (VEC) is increasingly recognized as essential for managing intensive computational demands in connected and autonomous vehicles. Due to limited onboard processing capabilities, vehicles must offload computational tasks to edge computing nodes, such as Road Side Units, to meet stringent latency and energy efficiency requirements. However, the inherent dynamic mobility of vehicles, intermittent connectivity, and fluctuating resource availability pose significant challenges to efficient task scheduling and offloading. Traditional static or heuristic-based approaches frequently result in deadline violations and inefficient energy usage under such dynamic conditions. This paper proposes an adaptive task scheduling framework using Deep Reinforcement Learning (DRL) to address these challenges. Our framework dynamically manages computational offloading decisions, optimizing for minimal deadline violations and reduced energy consumption. By integrating real-time state information, including server loads, network bandwidth, and task parameters, the DRL agent autonomously learns optimal scheduling policies through interaction with the vehicular environment. Our results indicate that this DRL-based scheduling significantly enhances adherence to task deadlines and improves energy efficiency. Furthermore, this foundational framework sets the stage for future expansions to include comprehensive Quality of Service (QoS) metrics and Service Level Agreement (SLA) compliance.

Microservice architectures are increasingly used to modularize IoT applications and deploy them in distributed and heterogeneous edge computing environments. Over time, these microservice-based IoT applications are susceptible to performance anomalies caused by resource hogging (e.g., CPU or memory), resource contention, etc., which can negatively impact their Quality of Service and violate their Service Level Agreements. Existing research on performance anomaly detection for edge computing environments focuses on model training approaches that either achieve high accuracy at the expense of a time-consuming and resource-intensive training process or prioritize training efficiency at the cost of lower accuracy. To address this gap, while considering the resource constraints and the large number of devices in modern edge platforms, we propose two clustering-based model training approaches: (1) intra-cluster parameter transfer learning (ICPTL)-based model training and (2) cluster-level model (CM) training. These approaches aim to find a tradeoff between the training efficiency of anomaly detection models and their accuracy. We compared the models trained under ICPTL and CM to models trained for specific devices (most accurate, least efficient) and a single general model trained for all devices (least accurate, most efficient). Our findings show that ICPTL’s model accuracy is comparable to that of the model per device approach while requiring only 40% of the training time. In addition, CM further improves training efficiency by requiring 23% less training time and reducing the number of trained models by approximately 66% compared to ICPTL, yet achieving a higher accuracy than a single general model.

The Internet of Things (IoT) revolutionizes smart city domains such as healthcare, transportation, industry, and education. The Internet of Medical Things (IoMT) is gaining prominence, particularly in smart hospitals and Remote Patient Monitoring (RPM). The vast volume of data generated by IoMT devices should be analyzed in real-time for health surveillance, prognosis, and prediction of diseases. Current approaches relying on Cloud computing to provide the necessary computing and storage capabilities do not scale for these latency-sensitive applications. Edge computing emerges as a solution by bringing cloud services closer to IoMT devices. This paper introduces SmartEdge, an AI-powered smart healthcare end-to-end integrated edge and cloud computing system for diabetes prediction. This work addresses latency concerns and demonstrates the efficacy of edge resources in healthcare applications within an end-to-end system. The system leverages various risk factors for diabetes prediction. We propose an Edge and Cloud-enabled framework to deploy the proposed diabetes prediction models on various configurations using edge nodes and main cloud servers. Performance metrics are evaluated using, latency, accuracy, and response time. By using ensemble machine learning voting algorithms we can improve the prediction accuracy by 5% versus a single model prediction.

ChatGPT is a language model based on Generative AI. Existing research work on ChatGPT focused on its use in various domains. However, its potential for Sign Language Translation (SLT) is yet to be explored. This paper addresses this void. Therefore, we present GPT's evolution aiming a retrospective analysis of the improvements to its architecture for SLT. We explore ChatGPT's capabilities in translating different sign languages in paving the way to better accessibility for deaf and hard-of-hearing community. Our experimental results indicate that ChatGPT can accurately translate from English to American (ASL), Australian (AUSLAN), and British (BSL) sign languages and from Arabic Sign Language (ArSL) to English with only one prompt iteration. However, the model failed to translate from Arabic to ArSL and ASL, AUSLAN, and BSL to Arabic. Consequently, we present challenges and derive insights for future research directions.

Vehicular Ad Hoc Networks (VANET) is an emerging technology that enables a comfortable, safe, and efficient travel experience by providing mechanisms to execute applications related to traffic congestions, road accidents, autonomous driving, and entertainment. The mobile vehicles in VANET are characterized by low computational and storage capabilities. In such scenarios, to meet applications’ performance requirements, requests from vehicles are offloaded to edge and cloud servers. The high energy consumption of these servers increases operating costs and threatens the environment. Energy-aware offloading strategies have been introduced to tackle this problem. Existing works on computation offloading focus on optimizing the energy consumption of either the IoT devices/mobile/vehicles and/or the edge servers. This paper proposes a novel offloading algorithm that optimizes the energy of edge–cloud integrated computing platforms based on Evolutionary Genetic Algorithm (EGA) while maintaining applications’ Service Level Agreement (SLA). The proposed algorithm employs an adaptive penalty function to incorporate the optimization constraints within EGA. Comparative analysis and numerical experiments are carried out between the proposed algorithm, random and genetic algorithm-based offloading, and no offloading baseline approaches. On average, the results show that the proposed algorithm saves 2.97 times and 1.37 times more energy than the random and no offloading algorithms respectively. Our algorithm has 0.3% of violations versus 52.8% and 62.8% by the random and no offloading approaches respectively. While the energy-non-SLA-aware genetic algorithm saves, on average, 1.22 times more energy than our approach, however, it violates SLAs by 159 times more than our proposed approach.

Over the last decade, the management of healthcare records has been revolutionized due to the need for accurate and cost-efficient patient-centric care alongside the technological advances. Currently, Electronic Health Records (EHRs) are managed using a client-server architecture by which healthcare providers retain the data stewardship. However, this approach suffers from security and privacy issues, a single point of failure, data fragmentation and vulnerability. The data replication, immutability, transparency, security and privacy features of blockchain have a promising future in the healthcare domain addressing the existing issues. In this paper, we propose BlockHR, a healthcare records management framework for healthcare providers and patients enabling better prognosis/diagnosis and follow-up. We analyze the effectiveness of BlockHR in providing security and privacy compared to the client-server approach. We also evaluate the performance of BlockHR versus the client-server approach. Our experimental results demonstrate that client-server approach takes 2.6 times less execution time for data write operation compared to BlockHR. The data retrieval for BlockHR is 20 times faster compared to the client-server approach. We analyze the impact of increasing number of medical records in our evaluation.

Cloud Internet of Things (IoT) is a novel paradigm, where the limitations of IoT associated devices in terms of storage, data access, scalability, networking and computing, and complex analysis are solved through use of the cloud computing infrastructure. The pervasive adoption of cloud in the IoT framework, makes the underlying data centers exacerbate problems like the environmental carbon footprint and operational costs which arise from the high energy consumption of computing servers. Several works proposed virtual machine placement and task scheduling algorithms to reduce the energy consumption of the underlying cloud infrastructure. However, each algorithm uses a different environment, experimental setup, power consumption model and workload for its evaluation, making it difficult to compare among them. In this paper, we give a classification and evaluation of 13 different algorithms using a unified setup, with the aim of achieving an objective comparison. The workload used for the evaluation is selected to typify IoT applications, such as connected vehicles, wide area measurement systems for the power grid, and smart meters for advanced meter infrastructure. The detailed performance analysis is elaborated in this paper.

Summary Cloud computing is an emerging technology in which information technology resources are virtualized to users in a set of computing resources on a pay‐per‐use basis. It is seen as an effective infrastructure for high performance applications. Divisible load applications occur in many scientific and engineering applications. However, dividing an application and deploying it in a cloud computing environment face challenges to obtain an optimal performance due to the overheads introduced by the cloud virtualization and the supporting cloud middleware. Therefore, we provide results of series of extensive experiments in scheduling divisible load application in a Cloud environment to decrease the overall application execution time considering the cloud networking and computing capacities presented to the application's user. We experiment with real applications within the Amazon cloud computing environment. Our extensive experiments analyze the reasons of the discrepancies between a theoretical model and the reality and propose adequate solutions. These discrepancies are due to three factors: the network behavior, the application behavior and the cloud computing virtualization. Our results show that applying the algorithm result in a maximum ratio of 1.41 of the measured normalized makespan versus the ideal makespan for application in which the communication to computation ratio is big. They show that the algorithm is effective for those applications in a heterogeneous setting reaching a ratio of 1.28 for large data sets. For application following the ensemble clustering model in which the computation to communication ratio is big and variable, we obtained a maximum ratio of 4.7 for large data set and a ratio of 2.11 for small data set. Applying the algorithm also results in an important speedup. These results are revealing for the type of applications we consider under experiments. The experiments also reveal the impact of the choice of the platforms provided by Amazon on the performance of the applications under study. Considering the emergence of cloud computing for high performance applications, the results in this paper can be widely adopted by cloud computing developers. Copyright © 2014 John Wiley & Sons, Ltd.

SUMMARY Cloud computing is an emerging technology where information technology resources are provisioned to users in a set of a unified computing resources on a pay per use basis. The resources are dynamically chosen to satisfy a user service level agreement and a required level of performance. A cloud is seen as a computing platform for heavy load applications. Conjugate gradient (CG) method is an iterative linear solver that is used by many scientific and engineering applications to solve a linear system of algebraic equations. CG generates a heavy load of computation, and therefore, it slows the performance of the applications using it. Distributing CG is considered as a way to increase its performance. However, running a distributed CG, based on a standard API, such as Message Passing Interface, in a cloud face many challenges, such as the cloud processing and networking capabilities. In this work, we present an in‐depth analysis of the CG algorithm and its complexity to develop adequate distributed algorithms. The implementation of these algorithms and their evaluation in our cloud environment reveal the gains and losses achieved by distributing the CG. The performance results show that despite the complexity of the CG processing and communication, a speedup gain of at least 1157.7 is obtained using 128 cores compared with National Aeronautics and Space Administration Advanced Supercomputing sequential execution. Given the emergence of clouds, the results in this paper analyzes performance issues when a generic public cloud, along with a standard development library, such as Message Passing Interface, is used for high‐performance applications, without the need of some specialized hardware and software. Copyright © 2012 John Wiley & Sons, Ltd.

Convolution represents a major computational load for many scientific and engineering applications, including seismic surface simulations and seismic imaging. Since convolution presents a heavy computational load, increasing its efficiency can significantly enhance the performance of associated applications. In this work, we present an in-depth analysis of the convolution algorithm and its complexity in order to develop adequate parallel algorithms. The implementation of these algorithms and their evaluation on the IBM Cell Broadband Engine (BE) processor reveals the gains and losses achieved by parallelizing the direct convolution. The performance results show that despite the complexity of the convolution processing, a speedup gain of at least 71.4 is obtained. The parallel vectorized algorithm requires the development effort of considering three independent vectorization strategies. Given the wide availability of Cell processors, the proposed parallelization approach can be widely adopted by any convolution-based application.

This paper describes authentication mechanisms for mobile agents. In these mechanisms, the authentication of mobile agents is controlled by the mobile-agents platform using digital signature and a public key infrastructure. Agents are authenticated via the authentication of their running platforms. An important advantage of our technique is transparency of the mechanisms and the portability of non-secure applications onto a secure environment. The authentication mechanisms are integrated within the transportation layer of a mobile-agents system. A minimal mobile agent system and the authentication mechanisms have been implemented. The implementation experiments have shown the feasibility and the advantage of these mechanisms

This paper presents a performance evaluation of the mobile agent paradigm in comparison to the client/server paradigm. This evaluation has been conducted on top of the Java environment, using respectively RMI, the Aglets mobile agents platform and a mobile agents prototype that we implemented. The measurements give the cost of the basic mechanisms involved in the implementation of a mobile agent platform, and a comparative evaluation of the two considered models (client/server and mobile agents) through two application scenarios. The results show that significant performance improvements can be obtained using mobile agents.

: This paper describes a protection scheme for mobile agents implemented on a Java environment. In this scheme, access to objects is controlled by means of software capabilities that can be exchanged between mutually suspicious agents. Each agent defines the access control rules that must be enforced when interacting with other agents. An important advantage of the proposed scheme is that the definition of the protection policy of an agent (i.e. how access rights are granted to other agents) is completely separated from the application code of that agent. It is described in an extended Interface Definition Language (IDL) at the interface level, thus enforcing modularity and ease of expression. A prototype has been implemented and experiments with simple agent-based applications have shown the feasibility and the advantage of this method. 1 Introduction Protection is a crucial aspect of distributed computing, in particular when users co-operate using shared objects or shared programs. ...