Data placement in dynamic fog ecosystems

Abstract

Dynamic data placement in distributed fog databases needs to employ different mechanisms than those typically encountered in modern cloud-hosted storage solutions to account for node instability and latency in receiving the results from queries running across multiple sites. In this work, we examine two dynamic data placement policies. The first is based on the fog node stability metadata, while the second one is driven by analytic applications running on top of distributed storages taking into account the latency, data freshness and quality objectives. Both our policies are enabled through extending Apache Ignite.