Traffic prediction and load balancing based on in-band network telemetry data

To address the challenges of massive telemetry data analysis in network operation and management, a network control automation system was designed. The overall architecture followed the “monitor-analyze-control” principle. The SRv6-INTO algorithm was used in the monitoring module for telemetry data...

Full description

Saved in:
Bibliographic Details
Main Authors: ZHANG Xu, ZHAO Chenxi, FENG Chuan, WEN Ting, LIU Wei, GUO Lei
Format: Article
Language:zho
Published: Editorial Department of Journal on Communications 2025-01-01
Series:Tongxin xuebao
Subjects:
Online Access:http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2025142/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:To address the challenges of massive telemetry data analysis in network operation and management, a network control automation system was designed. The overall architecture followed the “monitor-analyze-control” principle. The SRv6-INTO algorithm was used in the monitoring module for telemetry data collection and orchestration. To alleviate the load on the control plane, a data processor was developed in the data plane to perform preliminary processing of the telemetry data. The analysis module employed a combination of long short-term memory networks and attention mechanisms for traffic forecasting based on the telemetry data. The control module dynamically adjusted service flow paths using SRv6 technology, based on the prediction results from the analysis module. Simulation results show that the proposed model outperforms other methods in prediction accuracy and successfully achieves congestion avoidance.
ISSN:1000-436X