2019 IEEE Conference on Network Softwarization (NetSoft) 2019
DOI: 10.1109/netsoft.2019.8806669
|View full text |Cite
|
Sign up to set email alerts
|

Service Function Migration Scheduling based on Encoder-Decoder Recurrent Neural Network

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
5
0

Year Published

2020
2020
2023
2023

Publication Types

Select...
5
3
1

Relationship

0
9

Authors

Journals

citations
Cited by 12 publications
(5 citation statements)
references
References 10 publications
0
5
0
Order By: Relevance
“…The resource control system inds a solution for resource arbitration among CNFs or NF migration between physical servers by using the CPU utilization prediction results received from the advanced data analytics system, and/or other external information such as network and service requirements. Because the main contribution of this paper is an advanced data analytics system, we omit a detailed explanation of the structure and mechanisms of the resource control system and refer readers to [22]- [26].…”
Section: Resource Control Systemmentioning
confidence: 99%
See 1 more Smart Citation
“…The resource control system inds a solution for resource arbitration among CNFs or NF migration between physical servers by using the CPU utilization prediction results received from the advanced data analytics system, and/or other external information such as network and service requirements. Because the main contribution of this paper is an advanced data analytics system, we omit a detailed explanation of the structure and mechanisms of the resource control system and refer readers to [22]- [26].…”
Section: Resource Control Systemmentioning
confidence: 99%
“…The application of ML techniques to data analytics contributes to satisfying various levels of QoS requirements in 5G/6G network and cloud systems based on their capability to enhance computational resource control and management by autonomously suppressing resource shortage with high ef iciency of resource utilization, as discussed in [22]- [27]. Such enhancement can also be achieved for the resource control and management of microservices (such as CNFs) by exploiting the advantages of ML techniques, such as reducing the time required to ind an appropriate solution for computational resource adjustments compared to existing complex optimization approaches (as discussed in [26]), and enabling proactive resource control through training and prediction processes. In this paper, we irst introduce a framework of computational resource control and management for CNFs to provide microservice-based application services.…”
Section: Introductionmentioning
confidence: 99%
“…3) VNF Migration: The authors in [24] considered VNF migration scheduling problem based on an encoder-decoder recurrent neural network. In [25], the authors considered impact of user's mobility on VNF placement in a cellular network, aiming to minimize the cost of VNF migration.…”
Section: A Motivationmentioning
confidence: 99%
“…Khai et al formulate a MILP optimization problem with the objective of minimizing the migration time as part of their flexible, multi-step approach for VM migration [8]. Hirayama et al formulate an ILP problem to seek optimal resource allocation and train an encoder-decoder recurrent neural network, which is effective in preventing resource shortage and minimizing occurrences of VNF migration [9]. The works outlined above all address the VNF migration problem; however, they do not consider SLA requirements related to availability.…”
Section: Related Workmentioning
confidence: 99%