Virtual Full Replication: Achieving Scalability in Distributed Real-Time Main-Memory Systems

To achieve better scalability in a fully replicated distributed main-memory database, we propose support for virtual full replication. Full replication is often necessary for availability and predictability in critical embedded applications. In the fully replicated database, however, all updates are sent to all nodes, regardless of whether the data is ever used at every node. Virtual full replication is a concept that improves scalability without changing the application 's assumption of having access to a fully replicated database. We support virtual full replication by segmenting the database and allowing segments to have individual degrees of replication. This decreases the replication effort, lowers overall memory requirements for data and decreases node recovery time. Typical scenarios include distributed databases with many nodes, where often only a small number of the nodes need to share the same subset of information. We have previously specified a segmentation syntax for specifying important application semantics and outlined an implementation. Here, we analyze the potential scalability improvements in such an architecture