Monitoring distributed systems
ACM Transactions on Computer Systems (TOCS)
Tools for Distributed Application Management
Computer - Distributed computing systems: separate resources acting as one
Network and distributed systems management
Network and distributed systems management
Adaptive object-oriented filtering framework for event management applications
ACM Computing Surveys (CSUR)
On-Line Monitoring: A Tutorial
Computer
Application-Layer Group Communication Server for Extending Reliable Multicast Protocols Services
ICNP '97 Proceedings of the 1997 International Conference on Network Protocols (ICNP '97)
HiFi: A New Monitoring Architecture for Distributed Systems Management
ICDCS '99 Proceedings of the 19th IEEE International Conference on Distributed Computing Systems
Network management by delegation: the MAD approach
CASCON '91 Proceedings of the 1991 conference of the Centre for Advanced Studies on Collaborative research
High speed and robust event correlation
IEEE Communications Magazine
A survey of active network research
IEEE Communications Magazine
An architecture for monitoring, visualization, and control of gigabit networks
IEEE Network: The Magazine of Global Internetworking
The Tempest-a practical framework for network programmability
IEEE Network: The Magazine of Global Internetworking
Hi-index | 0.00 |
The successful deployment of next-generation distributed systems is significantly dependent on the efficient management support that improves the performance and reliability of these applications at run-time. This paper motivates and describes a programmable agents approach for active monitoring as an important attribute for supporting scalable, highly-responsive and non-intrusive management architecture. Active monitoring enables defining re-configurable and self-directed management tasks that can be modified automatically at run-time in order to track the system behavior. Based on observed events and users' monitoring demands, monitoring agents can dynamically customize their assigned tasks and initiate the appropriate monitoring actions. This avoids activating unnecessary monitoring tasks and provides a dynamic monitoring operations. The presented system, which is referred to as HiFi, supports a comprehensive environment including code instrumentation, user subscription, event filtering and action service. The paper also shows monitoring examples that illustrates the application and the effectiveness of active monitoring in managing large-scale distributed systems.