P2P data stores excel if availability of inserted data items must be guaranteed. Their inherent mechanisms to counter peer population dynamics make them suitable for a wide range of application domains. This paper presents and analyzes the fusion maintenance operation. It aims at reorganizing parts of our P2P data store in case the peer population shrinks so much that data availability is threatened. To this end, we present a formal cost model that peers use to estimate the optimal invocation point of a fusion. Finally, we present experimental results that validate our cost model by simulating various network conditions.