英语翻译5.4.Application Delegated ManagerThe Application Delegat

英语翻译5.4.Application Delegated ManagerThe Application Delegated Manager plays the role of abroker between components and resources.In addition,the ADM will have the responsibility of supervising theapplication execution at runtime to make sure we achieveall the required autonomic properties identified in theapplication service template as discussed in theApplication Management Editor subsection.The mainfbnctions of ADM can be outlined as follows:Maintain an up-to-date list of resources as well asavailable components that can participate in executinga given application.Select the set of resources that can meet theautonomic requirements of an application.0 Dispatch the appropriate number of mobile agents todeploy,configure,perform on-line monitoring,analysis,and adaptation when the application cannotmeet its requirements.5.5.Fault handlerThere are three types of fault handlers:System FaultHandler,Component Fault Handler and Agent FaultHandler.They deal with system fault,component faultand agent faults,respectively.These handlers are createdby the ADM and they share the component deploymentinformation.The ADM registers with the CoordinateSpace (CS) that it is interested in the presence of the faultentries in the Coordinate Space.The fault handlers aredesignated as event listeners once the fault entries appear.In the next subsections,we discuss the fault-detection andrecovery schemes adopted to achieve self-healing.Wealso show the dumpscreens of our proof-of-conceptprototype implemented based on Autonomia architecture.5.5.1.Fault Detection and Recovery Scheme.Thisimplementation has the capability to handle three differentkinds of faults defined as Component fault,Agent Systemfault and Node fault.In our environment,the ADMcreates a Node Fault Handler,Component Fault Handler,and Agent Fault Handler,which register with theJavaSpaces their interest in the presence of the node faultentry,the component fault entry and the agent fault entry,respectively.The JavaSpaces will notify the appropriatefault handler when a fault entry is written in the space.Then,the fault handler will spawn a thread to recover thatfault using checkpointinglmigrating mechanism [25].In Autonomia,whenever a fault is detected,thecorresponding fault entry is created and is then written tothe JavaSpaces.The lease of staying in the JavaSpaces forthe entry is renewed by a global Lease Renewal Managerservice.Any fault entry should stay in the JavaSpacesuntil a fault handler takes it out and deals with the faulteven after their creators have crashed.文章依旧很长,一定要尽量准确哦,