Junwei Cao Darren J. Kerbyson Graham R. Nudd Use of Agent-Based Service Discovery for Resource Management in Metacomputing Environment Junwei Cao Darren J. Kerbyson Graham R. Nudd Department of Computer Science University of Warwick
PACE Toolkit User Interface Resource Tools Application Tools Code Analysis Object Editor Object Library CPU Network Cache HMCL Scripts Resource Model Resource Model PSL Scripts Evaluation Engine Compiler Application Model Application Model Evaluation Engine Performance Analysis Performance Analysis On-the-fly Analysis
The Question Is … ?
A4 Methodology An agent is … A local manager An user middleman A broker A coordinator A service provider A service requestor A matchmaker A router
Service Discovery NEXT! Service Advertisement Local Management Layer Coordination Layer Communication Layer Service Advertisement NEXT!
Service Advertisement Hi, please find attached my service information. Hi, could you please give me some service information that you have? Full service advertisement – requires no service discovery. No service advertisement – results in complex service discovery. Make Balance!
Agent Capability Tables The process of the service advertisement and discovery corresponds to the maintenance and lookup of the ACTs. Vary by source: T_ACT: contains service info of local resources L_ACT: contains service info coming from lower agents G_ACT: contains service info coming from upper agent C_ACT: contains cached service info during discovery Strategies: Data-push: submit service info to other agents Data-pull: ask for service info from other agents Periodical: Periodical ACT maintenance Event-driven: ACT maintenance driven by system events
The Answer Is … At meta level, At local level, agents cooperate with each other for service discovery. At local level, PACE functions can supply accurate performance info.
ARMS ARMS in Context A4 Grid Grid Users Resources PACE A4 Simulator PMA PACE Application Tools (AT) Evaluation Engine (EE) Resource Tools (RT)
ARMS Architecture ? Bottleneck? ACT EE Agents PMA Application Models Resource Models ? Application Models Cost Models Users AT RT Processors
Application Execution ARMS Agent Structure Local Application Management Resource Allocation Monitoring Application Execution Coordination Sched. Cost ACTs App. Info Service Info Res. Info Scheduler Match Maker Eval Results Cost Model Advertisement ACT Manager PACE Evaluation Engine To Another Agent Agent ID Application Model Comm. Communication Module Discovery
PMA Structure ARMS Agent PMA Model Composer Monitoring Reconfiguration Statistical data Model Composer Simulation Engine Policies Performance Model Strategies
Conclusions Performance prediction driven for QoS support of grid resource management Agent based hierarchical model for grid resource advertisement and discovery Simulation based performance optimisation and steering of service discovery in large scale multi-agent systems In summary, all of above go together to provides an available methodology and prototype implementation of agent-based resource management for grid computing, which can be used as a fundamental framework for further improvement and refinement.