A Task Offloading Algorithm With Cloud Edge Jointly Load Balance Optimization Based on Deep Reinforcement Learning for Unmanned Surface Vehicles

Yan, Linjie; Chen, Haiming; Tu, Youpeng; Zhou, Xinyan

doi:10.1109/access.2022.3150406

Cited by 11 publications

(9 citation statements)

References 20 publications

(26 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed LBEEMM model addresses the limitations of IDS and Blockchain-based security models by incorporating a Deep Reinforcement Learning-based Iterative-learning Contextual Side chaining Model process. This innovative approach not only learns and adapts to new security threats over time but also uses contextual side-chaining to link related security events, thereby providing a robust and dynamic defences mechanism against potential threats [33,34,35]. Load balancing is a critical aspect of distributed computing systems, ensuring the efficient utilization of resources and optimal performance across heterogeneous environments.…”

Section: Literature Surveymentioning

confidence: 99%

VMMISD: An Efficient Load Balancing Model for Virtual Machine Migrations via Fused Metaheuristics With Iterative Security Measures and Deep Learning Optimizations

Brahmam,

2024

IEEE Access

View full text Add to dashboard Cite

As cloud computing continues to evolve, the need for efficient and secure management of virtual machine (VM) migrations has become increasingly evident. Traditional models often fall short in optimizing load balancing and energy consumption while ensuring a high level of security. In this work, we propose the Load Balancing and Energy-Efficient Migration Model, an innovative approach that leverages bioinspired algorithms and advanced security measures to optimize VM migrations. The initial novelty of our model is the integration of Genetic Algorithms with Ant Colony Optimization for resource scheduling operations. These algorithms were specifically chosen for their proven effectiveness in solving complex optimization problems by simulating natural processes. Additionally, our model incorporates a Deep Reinforcement Learning-based Iterative-learning Contextual Side chaining Model to enhance security measures. This approach not only learns and adapts to new security threats over time but also utilizes contextual side-chaining to link related security events, thereby providing a robust defense mechanism against potential threats. The affinity between VMs and physical machines is quantified using K Means Clustering and Fuzzy Logic, which ensures optimal load balancing while accounting for the uncertainty inherent in the migration process. Furthermore, we employ Bidirectional Long Short-Term Memory networks with Recurrent Graph Neural Network, for accurate workload prediction and informed migration decision Making process. The selection of these techniques is grounded in their proven capability to analyze historical data and predict future trends with high accuracy levels. Our proposed model demonstrates marked improvements in several key performance metrics. We achieved a 4.5% reduction in makespan, a 4.9% increase in deadline hit ratio, and a 3.9% improvement in task diversity. Furthermore, computational complexity was reduced by 8.3%, VM migration efficiency improved by 2.5%, and the delay of computation was significantly reduced by 9.5%. Importantly, the integration of the Iterative-learning Contextual Side chaining Model significantly enhanced the security and quality of service (QoS) under attack scenarios, resulting in a 10.4% improvement in response speed, a 2.5% reduction in energy consumption during block mining, a 3.9% improvement in throughput, and an 8.5% reduction in storage costs. This Load Balancing and Energy-Efficient Migration Model represents a significant advancement in addressing the challenges of load balancing, energy efficiency, and security in VM migrations. Through the meticulous integration of bioinspired algorithms, advanced security measures, and machine learning techniques, our model provides a comprehensive and innovative solution that markedly improves system performance, reduces energy consumption, and fortifies security, thereby paving the way for a more efficient and secure cloud computing ecosystem.

show abstract

Section: Literature Surveymentioning

confidence: 99%

VMMISD: An Efficient Load Balancing Model for Virtual Machine Migrations via Fused Metaheuristics With Iterative Security Measures and Deep Learning Optimizations

Brahmam,

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…In this study, we employ deep Q-network (DQN) as a fundamental and representative DRL algorithm and enhance its training process by developing a feature map and DNN structure. Importantly, these developed components can be reused for evaluating the state in other advanced DRL algorithms, such as double DQN (DDQN) [41] and deep deterministic policy gradient (DDPG) [31], [42], [43]. At every time step t, the average return, called Q-value, is computed to evaluate state-action pairs and strategy is updated on the basis of the Q-value.…”

Section: B Deep Reinforcement Learningmentioning

confidence: 99%

Learning-Based Joint User Association and Cache Replacement for Cache-Enabled Cloud RAN

Jeon,

Jung,

Lee

et al. 2024

IEEE Open J. Commun. Soc.

View full text Add to dashboard Cite

Mobile edge caching is regarded as a promising technology for reducing network latency and alleviating network congestion by efficiently offloading data traffic and computations to cache-enabled edge nodes. To fully leverage the benefits of edge caching, it is essential to jointly optimize caching and communication strategies, accounting for dynamic content request pattern and unstable nature of wireless mobile networks. Motivated by this, we study a joint cache replacement and user association strategy for minimizing the content delivery latency in cache-enabled cloud radio access network (C-RAN) where remote radio heads (RRHs) cache some contents for serving the content request without downloading the requested content from centralized baseband unit (BBU) via fronthaul. Unlike traditional cache placement strategies, our cache replacement facilitates gradual and timely updates while serving user content requests, without imposing additional network overhead. Specifically, whenever a user requests a content, BBU makes decisions on selecting a RRH for serving user request and on replacing the cached data of the selected RRH by taking into account the user location, cache status of RRHs, and impact on subsequent content deliveries. We optimize the selection of RRH to serve user request and the replacement of cached data by formulating a latency minimization problem using Markov Decision Process (MDP). This formulation considers the tradeoff between cache hit ratio and communication reliability. To develop an effective strategy for solving the MDP, we employ a deep reinforcement learning (DRL) algorithm and design a novel neural network structure and input feature map, specifically tailored to our problem domain. Simulation results show that the proposed approach learns effective strategy appropriate to a given environment, thereby outperforming not only the traditional rule-based strategies but also a typical DRL algorithm in terms of average latency. The proposed approach is shown to be relatively robust to time-variant content popularity by quickly adapting to new popularity distribution.

show abstract

“…Several centralized offloading algorithms are proposed in [17][18][19] to solve the above problems while considering the total system state. An online predictive offloading algorithm based on DRL and Long Short-Term Memory (LSTM) networks is proposed in [17], it predicts the load of the ES in real time during the model's training phase and allocates the computational resources for the task in advance to substantially increase the convergence speed and accuracy of the DRL algorithm during the offloading process.…”

Section: Related Workmentioning

confidence: 99%

“…An online predictive offloading algorithm based on DRL and Long Short-Term Memory (LSTM) networks is proposed in [17], it predicts the load of the ES in real time during the model's training phase and allocates the computational resources for the task in advance to substantially increase the convergence speed and accuracy of the DRL algorithm during the offloading process. In [18], a DRL-based Task Offloading with cloud edge jointly Load Balance Optimization (TOLBO) algorithm is proposed to select the best ES or CS for offloading in order to minimize long-term task latency and energy consumption by jointly considering the requirements of latency and energy-sensitive tasks and the overall load dynamics in the cloud, edge, and end layers. An advanced DRL-based offloading algorithm is proposed in [19], considering the previous processing time for MDs, ESs, and CS, it can generate and store multi-class offloading decisions with the system state together in a database and then training and updating multiple parallel CNNs with a batch of labeled data for executing independent tasks.…”

Section: Related Workmentioning

confidence: 99%

DRL-Based Distributed Task Offloading Framework in Edge-Cloud Environment

Nashaat,

Hashem,

Rizk

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The Internet of Things (IoT), real-time media streaming, and other related technologies have increased due to the rapid development of wireless communication technologies and the enormous growth of computation, storage, and data transmission tasks. Edge-Cloud Computing (ECC) is a technology that can be used to better meet the diverse needs of IoT users, it combines the benefits of Mobile Cloud Computing (MCC) and Mobile Edge Computing (MEC) to meet energy consumption and delay requirements, and achieve more stable and affordable task execution. The most significant challenge in ECC is making realtime task offloading decisions. In order to generate offloading decisions in ECC environments in an efficient and near optimal manner, a Deep Reinforcement Learning (DRL)-based Distributed task Offloading (DRL-DO) framework is proposed. Simulation results demonstrate the accuracy of the DRL-DO framework, it achieves high Gain Ratio (GR) and greatly reduces the energy consumption, response time, while attaining moderate time cost compared with other offloading algorithms.

show abstract

A Task Offloading Algorithm With Cloud Edge Jointly Load Balance Optimization Based on Deep Reinforcement Learning for Unmanned Surface Vehicles

Cited by 11 publications

References 20 publications

VMMISD: An Efficient Load Balancing Model for Virtual Machine Migrations via Fused Metaheuristics With Iterative Security Measures and Deep Learning Optimizations

VMMISD: An Efficient Load Balancing Model for Virtual Machine Migrations via Fused Metaheuristics With Iterative Security Measures and Deep Learning Optimizations

Learning-Based Joint User Association and Cache Replacement for Cache-Enabled Cloud RAN

DRL-Based Distributed Task Offloading Framework in Edge-Cloud Environment

Contact Info

Product

Resources

About