Title :
Intelligent agents for proactive fault detection
Author :
Hood, Cynithia S. ; Ji, Chuanyi
Author_Institution :
Illinois Inst. of Technol., Chicago, IL, USA
Abstract :
As the Internet becomes a critical component of our society, a key challenge is to maintain network availability and reliability. Intelligent processing agents that reside at network nodes use an adaptive learning method to detect abnormal network behavior before a fault actually occurs. In a test at Rensselaer Polytechnic Institute, an agent on a router detected a file server failure 12 minutes before it occurred
Keywords :
Internet; diagnostic expert systems; fault diagnosis; learning (artificial intelligence); software agents; telecommunication computing; Internet; abnormal network behavior detection; adaptive learning method; file server failure; intelligent agents; network availability; network nodes; network reliability; proactive fault detection; router; Bayesian methods; Fault detection; Fault diagnosis; Hardware; Humans; IP networks; Intelligent agent; Learning systems; Risk management; Technology management;
Journal_Title :
Internet Computing, IEEE
DOI :
10.1109/4236.670686
