Implementation of Alarm Correlation System for Hybrid Networks Based upon the perfSONAR Framework

  • Authors:
  • Jen-Wei Hu;Hui-Min Chen;Te-Lung Liu;Hui-Min Tseng;Daniel Lin;Chu-Sing Yang;C. Eugene Yeh

  • Affiliations:
  • -;-;-;-;-;-;-

  • Venue:
  • WAINA '10 Proceedings of the 2010 IEEE 24th International Conference on Advanced Information Networking and Applications Workshops
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Based on services-oriented architecture(SOA), perfSONAR is a web service infrastructure. The set of elementary functions have been isolated and can be provided by different services. A primary goal of perfSONAR is to make it easier to detect performance monitoring problems among different network domains. perfSONAR has been widely accepted as one of the web service standards by NRENs (National Research and Education Network). The major services are Lookup, BandWidth ConTroLer(BWCTL), PingER and so on. TWAREN(TaiWan Advanced Research and Education Network) is deployed as a hybrid network that employs both IP and Optical technologies. In order to locate the root-cause as alarms occurring, we proposed an alarm correlation algorithm. In this paper, we design a novel system based on our previous work which provides an alarm analysis service in hybrid networks according to the perfSONAR framework. In this system, we developed TL1, SNMPLite and Correlation Web Services that retrieve alarm messages from optical facilities, routers and locate the root cause respectively. Finally, analysis result is presented by PerfSONARUI. The system inherits SOA and cross-platform characteristics of perfSONAR. With this system, we will provide TWAREN NOC a faster and more precise approach to locate the root cause of network problems. It offers engineers an assistance to reduce failure eliminating time and enhance network availability.