AbstractWe consider the problem of deploying and managing replicated highly available services in federated systems that spanmultiple collaborative organizations. In particular, we present a peer-to-peer framework that supports the construction of self-managing replicated services that automatically adjust the number of replicas and their placement in response to changes in the system or client loads. Our framework is novel in that self-management is completely decentralized, depending only on a modest amount of loosely synchronized global state. More specifically, each replica is wrapped with a management agent that monitors the state of the service and the underlying system. Each agent occasionally searches for configurations that may be better than the current one with respect to an application model and explicitly stated performance and availability targets. On finding a better configuration, an agent will enact the new configuration after a random delay to avoid possible collisions. We evaluate our framework by studying a prototype UDDI service subjected to various system changes as well as client load. We show that while agents act autonomously, the service rapidly reaches a stable and appropriate configuration in response to system dynamics.
RightsThis Item is protected by copyright and/or related rights.You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use.For other uses you need to obtain permission from the rights-holder(s).