Name
Abstract | ||
|---|---|---|
In this paper we present the design and evaluation of asystem for deployinghighly availableandmigratableservicesin shared infrastructures, such as the PlanetLab, where the availableresource capacities at a nodecan fluctuate significantly. A migratableservicecanmonitorits operating conditions and autonomously relocate itself to another node when the available resource capacitiesatthe currentnode fallbelowcertainacceptablelimits. Weutilizethe autonomousmobile agent paradigm forbuildingsuchmigratableservices. Such agents can monitor their operatingconditionsandfollow various migration policies.We investigate here the mechanisms for servicerelocation,andclient-sideprotocolstoaccess migratoryservices. The ``blackout periods'',i.e. the time during which the clients are unable to access a migratingservice,needtobe minimized and kept within some tolerable limits for services required to be highly available. We first present the design ofa migratable service implemented using a mobile agent, and evaluateitsperformance in terms of theblackout periods andthe service agent's abilities to autonomously migratein the network. Wereplicateserviceagentstoreducetheblackoutperiods, and develop the coordination protocols for autonomous agent migration in a group ofserviceagents.Wealsopresenthere our work for monitoring PlanetLab nodes for their available resource capacities in orderto assist a migratory serviceinselectingatarget node for relocation. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1109/SRDS.2009.39 | SRDS |
Keywords | Field | DocType |
autonomous agent migration,available resource capacity,andthe service agent,weutilizethe autonomousmobile agent paradigm,wide-area shared computing environments,migratory serviceinselectingatarget node,resource-aware migratory services,available resource capacitiesatthe currentnode,planetlab node,mobile agent,operatingconditionsandfollow various migration policy,design ofa migratable service,data mining,protocols,servers,operant conditioning,resource allocation,mobile communication,autonomous agent,fault tolerance | PlanetLab,Autonomous agent,Computer science,Mobile agent,Server,Computer network,Real-time computing,Fault tolerance,Resource allocation,Blackout,Mobile telephony,Distributed computing | Conference |
ISSN | Citations | PageRank |
1060-9857 | 3 | 0.40 |
References | Authors | |
17 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Anand Tripathi | 1 | 1151 | 106.92 |
| Vinit Padhye | 2 | 18 | 3.10 |
| Devdatta Kulkarni | 3 | 186 | 9.72 |