Ever-increasing scales of corporate networks configured as independently co
ntrolled sections having equal authority make it more and more difficult to
maintain conformity of the configuration. The authors investigated methods
of efficient implementation of network configuration modifications underta
ken in large-scale networks by managing the network configuration versions
using distributed database storing the configurational information for all
management domains. In this article, we propose an inter-agent protocol nec
essary for consistent operation of the entire network consisting of domains
managed by a management agent and identify conditions necessary for its im
plementation. We also provide data concerning the upper limits of the recov
ery time in the event of a failure as a result of network configuration man
agement and propose an algorithm showing that the upper limits of the failu
re recovery time increase linearly with an increase in the network scale. (
C) 2000 Scripta Technica.