Delay-Aware NFV Resource Allocation with Deep Reinforcement Learning

Yuan, Ningcheng; He, Wenchen; Shen, Jing; Qiu, Xuesong; Guo, Shaoyong; Li, Wenjing

doi:10.1109/noms47738.2020.9110377

Cited by 12 publications

(11 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, the Deep-RL framework takes the required action

, which consists of bandwidth RA

and IoT devices scheduling 𝒪 at every state

. The state environment can be defined as

where

represents loss value,

is the speed of the CPU cycle frequency at edge servers,

is the data rate vector that can be achieved and regulated by the bandwidth allocation policy, and

represents the learning for IoT device [ 25 , 28 ]. To achieve high efficiency in the learning accuracy and sustainability of the DT-IoT, the agent continues to the next state and receives a reward immediately.…”

Section: Formulation Of the Communication Effectiveness Problem For D...mentioning

confidence: 99%

Sustainable Resource Allocation and Reduce Latency Based on Federated-Learning-Enabled Digital Twin in IoT Devices

Alhartomi,

Salh,

Audah

et al. 2023

Sensors

View full text Add to dashboard Cite

In this article, we utilize Digital Twins (DT) with edge networks using blockchain technology for reliable real-time data processing and provide a secure, scalable solution to bridge the gap between physical edge networks and digital systems. Then, we suggest a Federated Learning (FL) framework for collaborative computing that runs on a blockchain and is powered by the DT edge network. This framework increases data privacy while enhancing system security and reliability. The provision of sustainable Resource Allocation (RA) and ensure real-time data-processing interaction between Internet of Things (IoT) devices and edge servers depends on a balance between system latency and Energy Consumption (EC) based on the proposed DT-empowered Deep Reinforcement Learning (Deep-RL) agent. The Deep-RL agent evaluates the performance action based on RA actions in DT to distribute its bandwidth resources to IoT devices based on iteration and the actions taken to generate the best policy and enhance learning efficiency at every step. The simulation results show that the proposed Deep-RL-agent-based DT is able to exploit the best policy, select 47.5% of computing activities that are to be carried out locally with 1 MHz bandwidth and minimize the weighted cost of the transmission policy of edge-computing strategies.

show abstract

“…Furthermore, the Deep-RL framework takes the required action

, which consists of bandwidth RA

and IoT devices scheduling 𝒪 at every state

. The state environment can be defined as

where

represents loss value,

is the speed of the CPU cycle frequency at edge servers,

is the data rate vector that can be achieved and regulated by the bandwidth allocation policy, and

Section: Formulation Of the Communication Effectiveness Problem For D...mentioning

confidence: 99%

Sustainable Resource Allocation and Reduce Latency Based on Federated-Learning-Enabled Digital Twin in IoT Devices

Alhartomi,

Salh,

Audah

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…With that, a multi-service delay-based and QoS-aware scheduling scheme was proposed in the downlink MAC channel to maintain a minimum of the end-to-end delay under congested network scenarios. Furthermore, several efforts as in [18], [19], [29] introduced individual solutions that consider delay, transmission rate, and channel awareness in their scheduling decision. For example, the authors in [18] develop a delay-aware NFV radio resource allocation with deep reinforcement learning.…”

Section: Related Workmentioning

confidence: 99%

“…Furthermore, several efforts as in [18], [19], [29] introduced individual solutions that consider delay, transmission rate, and channel awareness in their scheduling decision. For example, the authors in [18] develop a delay-aware NFV radio resource allocation with deep reinforcement learning. This is considered a flexible allocation technique that is developed to fulfill E2E delay needs.…”

Section: Related Workmentioning

confidence: 99%

“…A bold observation on the current downlink resource allocation schemes in the literature, as surveyed by [11], indicated that the vast majority of the offered solutions distinguish the targeted problem as a single-dimension on a specific QoS criterion. For instance, throughput maximization (aiming the system capacity) as achieved in [12]- [15], delay minimization in [16]- [18], or fair service rate as in [19], [20] by evenly distributing the limited radio resources distributions over the different traffic classes. Nonetheless, the absence of multidimensional QoS provisioning guarantee in the competition among radio resource allocation schemes will be a futureproof solution for the eMBB use cases in 5G; in order to gracefully handle the diversity of applications interacted through the 5G user device.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Delay-Based Resource Allocation With Fairness Guarantee and Minimal Loss for eMBB in 5G Heterogeneous Networks

2022

View full text Add to dashboard Cite

Heterogeneity is defined as one of the core merits in the 5th Generation (5G) mobile systems. Such a trait is prominent in modern 5G traffic scenarios, i.e., enhanced Mobile Broadband (eMBB), where coarse Quality of Service (QoS) profiles need to be satisfied over the shared transmission channel. For that, Radio Resource Management (RRM) became a challenging problem that should be figured out based on the traffic scenario. Extensive efforts dwelled in the literature attempt to solve the resource allocation problem in the downlink channel by maximizing the gain for a certain aspect in the QoS profile. This, however, poses a challenge for 5G-related traffic scenarios such that multi-dimension QoS shall be attainable. Therefore, in this paper, a Delay-aware Resource Allocation for Guaranteed Fairness and minimal Loss (DRAGFL) is proposed to enhance the QoS for Real-Time (RT) services in eMBB 5G Heterogeneous Networks. The ultimate objective of DRAGFL is to maintain a steadily low latency for delay-sensitive flows without restraining the service fairness and the data loss in the overload states. DRAGFL enhances the efficiency of the MAC layer scheduler throughout three phases. Firstly, generating delay-derived weight matrix of the resources blocks to radio bearers from eMBB different classes. Then, greedy-based flows prioritization for Resource Blocks (RBs) assignment. Finally, the data rate of the RT and Non RT (NRT) flows is further emphasized through a channel-aware principle that is invoked conditioning to the remaining radio resources and available data to be scheduled in the interval. The performance evaluation using system-level simulations reveals prominent gains of DRAGFL in sustaining a steadily low end-to-end delay in the high congestion states. Moreover, it maintains ideal states of service fairness with low data loss for different traffic loads to indicate a promising MAC scheduler for the developing scenarios of 5G.

show abstract

“…In this paper, similar to [12,13,17,16,19,21,22,23], we consider the average end-to-end delay between the source and the destination of each service chain as the QoS metric. Threshold d th r indicates the utmost average tolerable end-to-end delay of SFC r.…”

Section: System Model and Problem Statementmentioning

confidence: 99%

End-to-End Delay Guaranteed SFC Deployment: A Multi-level Mapping Approach

Yaghoubpour,

Bakhshi,

Seifi

2021

Preprint

View full text Add to dashboard Cite

Network Function Virtualization (NFV) enables service providers to maximize the business profit via resource-efficient QoS provisioning for customer requested Service Function Chains (SFCs). In recent applications, end-to-end delay is one of the crucial QoS requirement needs to be guaranteed in SFC deployment. While it is considered in the literature, accurate and comprehensive modeling of the delay in the problem of maximizing the service provider's profit has not yet been addressed. In this paper, at first, the problem is formulated as a mixed integer convex programming model. Then, by decomposing the model, a two-level mapping algorithm is developed, that at the first level, maps the functions of the SFCs to virtual network function instances; and in the second level, deploys the instances in the physical network. Simulation results show that the proposed approach achieves a near optimal solution in comparison the performance bound obtained by the optimization model, and also superiors the existing solutions.

show abstract

Delay-Aware NFV Resource Allocation with Deep Reinforcement Learning

Cited by 12 publications

References 8 publications

Sustainable Resource Allocation and Reduce Latency Based on Federated-Learning-Enabled Digital Twin in IoT Devices

Sustainable Resource Allocation and Reduce Latency Based on Federated-Learning-Enabled Digital Twin in IoT Devices

Delay-Based Resource Allocation With Fairness Guarantee and Minimal Loss for eMBB in 5G Heterogeneous Networks

End-to-End Delay Guaranteed SFC Deployment: A Multi-level Mapping Approach

Contact Info

Product

Resources

About