2022
DOI: 10.1016/j.pmcj.2022.101722
|View full text |Cite
|
Sign up to set email alerts
|

Online machine learning for auto-scaling in the edge computing

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
5
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
3
3
1

Relationship

1
6

Authors

Journals

citations
Cited by 10 publications
(5 citation statements)
references
References 20 publications
0
5
0
Order By: Relevance
“…Therefore, the scaling of processing services must be dynamic. Recent work has studied online machine learning for autoscaling, such as the one in [ 150 ], in which the authors present an autoscaling subsystem for container-based processing services. However, it will be interesting and promising to design dynamic autoscaling to ensure the scalability of the system with high QoS performance.…”
Section: Open Issues and Future Directionsmentioning
confidence: 99%
“…Therefore, the scaling of processing services must be dynamic. Recent work has studied online machine learning for autoscaling, such as the one in [ 150 ], in which the authors present an autoscaling subsystem for container-based processing services. However, it will be interesting and promising to design dynamic autoscaling to ensure the scalability of the system with high QoS performance.…”
Section: Open Issues and Future Directionsmentioning
confidence: 99%
“…Further research is needed to develop dynamic and efficient models and algorithms for managing and scaling virtualized resources at the network edge. In our previous work [52], we introduced a MAPE-K-based architecture for scaling VNF in an EC environment. Our previous autoscaling combines proactive workload prediction with reactive adjustments when the prediction model falls short of the desired quality.…”
Section: Auto-scaling In Edge Computingmentioning
confidence: 99%
“…Considering the strategies used to calculate the number of resources to be scaled, we evaluated three approaches: Linear, Exponential, and Hybrid. Such strategies and parameters used in the auto-scaling are better described in [52]. It is worth mentioning that we only considered the hybrid strategy for the ensemble-based autoscaling solution.…”
Section: Evaluating the Ensemblementioning
confidence: 99%
See 2 more Smart Citations