Automated and distributed network service monitoring

Abstract

Decentralized architectures have been proposed using functional hierarchies of several middle-level managers with delegation of management activities. However, no management applications have been implemented and widely used that could sanction this approach for real use. Also, current frameworks or mechanisms in network management do not directly address automation. More efficient network management systems with higher levels of availability and automation are needed. In this paper we propose a design and specify an automated and distributed network services monitoring system for Internet services that should be capable to effectively gather and calculate relevant operational parameters that will be used to determine the availability level of a network or application service. Based on this availability level, the system recommends or automatically triggers a service reconfiguration or a total or partial network service replication.