Journal on Communications ›› 2019, Vol. 40 ›› Issue (11): 171-179.doi: 10.11959/j.issn.1000-436x.2019229

• Distinguished and Excellent Young Scholars • Previous Articles     Next Articles

DL-assisted programmable multilayer network application awareness system for IP-over-EON

Zuqing ZHU,Jiawei KONG,Bin NIU,Shaofei TANG,Hongqiang FANG,Siqi LIU   

  1. School of Information Science and Technology,University of Science and Technology of China,Hefei 230027,China
  • Revised:2019-10-17 Online:2019-11-25 Published:2019-12-06
  • Supported by:
    The National Natural Science Foundation of China(61871357);The National Natural Science Foundation of China(61771445);The National Natural Science Foundation of China(61701472)

Abstract:

In order to realize real-time and fine-granularity network monitoring and adjustment,and satisfy the specific QoS demands of various applications,a deep learning (DL) assisted programmable multilayer network application performance awareness system for IP-over-EON was proposed.The distributed network monitoring based on network application performance awareness was combined with centralized network management.The multilayer and fine-grained network monitoring was implemented by distributed network monitoring,and the data analysis through DL was performed.Experimental results indicate that by combining distributed and centralized processing seamlessly,the proposed network monitoring system can not only realize timely and automatic network control and management but also provide superior scalability.

Key words: deep learning, elastic optical network, multi-layer in-band network telemetry

CLC Number: 

No Suggested Reading articles found!