Group-Delay Aware Task Offloading with Service Replication for Scalable Mobile Edge Computing

Mohamed, Shimaa; Sorour, Sameh; Hassanein, Hossam S.

doi:10.1109/globecom42002.2020.9348241

Cited by 9 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, these strategies lengthen the overall VM transmission delay, which also lowers the user QoE. The authors in [21] have provided some approaches for VMs to ensure high reliability to minimize the difficulties of service replication. This method necessitates constant error detection, and it is recoverable by shifting to an available instance.…”

Section: Related Workmentioning

confidence: 99%

A Hyper-Heuristic Approach for Quality of Experience Aware Service Placement Scheme in 5G Mobile Edge Computing

Islam,

Ahammed,

Siddique

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The 5 th Generation (5G) Mobile Edge Computing (MEC) addresses the problem of high end-to-end delay experienced by traditional cloud computing users by ensuring fast accessible and reliable computing resources. However, the deployment of service instances in MEC resources requires migration due to user mobility. While Proactive Migration of service instances at multiple MECs increases users' Quality-of-Experience (QoE), Reactive Migration might reduce the deployment cost at the expense of user QoE. In this paper, we have developed a framework, that distributes service instances proactively among the Edge Nodes depending on user movement trajectories to ensure faster migration of the service instances and deliver higher QoE within minimum VNF deployment cost considering users' budgets. The aforementioned Proactive Service Placement (PSP) problem is formulated as a Multi-Objective Linear Programming (MOLP) that brings a trade-off between these two conflicting objectives, maximizing user QoE and lowering VNF deployment cost. For large networks, the PSP problem is proven to be an NP-hard problem. Thus, we have developed an artificial intelligence-based Hyper-heuristic algorithm for PSP, called HPSP, which can provide a high-performing solution within polynomial time. The HPSP exploits Tabu Search Optimization as a high-level meta-heuristic algorithm that selects one of the three lower-level metaheuristic algorithms-Golden Eagle Optimizer, Sine Cosine Optimization, and Jellyfish Search Optimization depending on the situation. The results of numerical analysis describe that the HPSP system outperforms the other state-of-the-art works in terms of user QoE, cost, and the ratio of proactive to reactive service placements.

show abstract

Section: Related Workmentioning

confidence: 99%

A Hyper-Heuristic Approach for Quality of Experience Aware Service Placement Scheme in 5G Mobile Edge Computing

Islam,

Ahammed,

Siddique

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…To cope with task processing delay, energy consumption, and monetary cost, it aimed for reducing the cost of resource utilization for a certain system capability. [18] provided a group‐delay aware task offloading and service replication scheme. Given with the buffer size, channel quality and transmission distance between MEC servers, it minimizes the service delay of all users while satisfying the application‐specific delay requirements.…”

Section: Related Workmentioning

confidence: 99%

“…Before assessing the relative performance of DQTM, we fine tune the weighting parameters of DQTM, that is the learning rate and batch size in the mini-batch gradient descent. After that, DQTM is compared with the local, random allocation, and minimized average service time schemes, which are respectively termed as local, random and MAST [18] for brevity. The local scheme disables any task migration, which is used as a comparison baseline to highlight the effect by task migration.…”

Section: Simulation Settingmentioning

confidence: 99%

“…Meanwhile, a referential value set of deep Q-learning coefficients is derived. Second, we compare DQTM with the local, random task and MAST [18] migration schemes. According to experimental measures by sensitivity to task arrival rate and the number of MEC servers, performance results manifest that DQTM can enhance the overall service efficiency and resource utilization against varied task workload in the multi-server MEC system.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Intelligent task migration with deep Qlearning in multi‐access edge computing

Huang

Lin

2021

IET Communications

View full text Add to dashboard Cite

Multi-access edge computing provides computation and network resources in proximity to user applications in mobile environments. Deploying edge servers in network boundary can not only offload the heavy task loading on the cloud, but also alleviate resource-limited capabilities of mobile devices. Rather than many stand-alone edge servers, the concept of multi-server edge computing is recently advocated to contend with the issues of system scalability and service quality against dynamic task workload. This study exploits collaborative computing resources and designs a task migration strategy for multiple edge servers in mobile networks. This study formulates a queueing optimization problem of minimizing the overall service time in a multi-server system. An intelligent task migration scheme is then developed using the deep reinforcement learning and Q-learning techniques. With a variety of numerical attributes derived from the queueing model, this intelligent scheme can arrange the task distribution among edge servers to enhance the task processing capability. Simulation-based results show that the proposed task migration scheme can sustain service efficiency and resource utilization, which is promising as compared with conventional designs without collaborative intelligence in mobile environments.

show abstract