TY - GEN
T1 - Using neural networks to identify control and management plane poison messages
AU - Du, Xiaojiang
AU - Shayman, Mark A.
AU - Skoog, Ronald
PY - 2003
Y1 - 2003
N2 - Poison message failure propagation is a mechanism that has been responsible for large scale failures in both telecommunications and IP networks: Some or all of the network elements have a software or protocol 'bug' that is activated on receipt of a certain network control/management message (the poison message). This activated 'bug' will cause the node to fail with some probability. If the network control or management is such that this message is persistently passed among the network nodes, and if the node failure probability is sufficiently high, large-scale instability can result. Identifying the responsible message type can permit filters to be configured to block poison message propagation, thereby preventing instability. Since message types have distinctive modes of propagation, the node failure pattern can provide valuable information to help identify the CUlprit message type. Through extensive simulations, we show that artificial neural networks are effective in isolating the responsible message type.
AB - Poison message failure propagation is a mechanism that has been responsible for large scale failures in both telecommunications and IP networks: Some or all of the network elements have a software or protocol 'bug' that is activated on receipt of a certain network control/management message (the poison message). This activated 'bug' will cause the node to fail with some probability. If the network control or management is such that this message is persistently passed among the network nodes, and if the node failure probability is sufficiently high, large-scale instability can result. Identifying the responsible message type can permit filters to be configured to block poison message propagation, thereby preventing instability. Since message types have distinctive modes of propagation, the node failure pattern can provide valuable information to help identify the CUlprit message type. Through extensive simulations, we show that artificial neural networks are effective in isolating the responsible message type.
KW - Fault management
KW - Neural network
KW - Node failure pattern
KW - Poison message
UR - http://www.scopus.com/inward/record.url?scp=84904330061&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84904330061&partnerID=8YFLogxK
U2 - 10.1007/978-0-387-35674-7
DO - 10.1007/978-0-387-35674-7
M3 - Conference contribution
AN - SCOPUS:84904330061
SN - 9781475755213
T3 - IFIP Advances in Information and Communication Technology
SP - 621
EP - 634
BT - Integrated Network Management VIII
T2 - IFIP/IEEE 8th International Symposium on Integrated Network Management, IM 2003
Y2 - 24 March 2003 through 28 March 2003
ER -