Scaling Off-the-Shelf Databases with Vela: An approach based on Virtualization and Replication

Off-the-Shelf (OTS), relational databases can be challenging when deployed in the cloud. Given their architectures, limitations arise regarding performance (because of virtualization), scalability (because of multi-tenancy), and elasticity (because existing engines cannot easily take advantage of the application migration possibilities of virtualized environments). As a result, many database engines tailored to cloud computing differ from conventional engines in functionality, consistency levels, support for queries or transactions, and even interfaces. Efficiently supporting Off-the-Shelf databases in the cloud would allow to port entire application stacks without rewriting them. In this paper we present a system that combines snapshot isolation (SI) replication with virtualization to provide a flexible solution for running unmodified databases in the cloud while taking advantage of the opportunities cloud architectures provide. Unlike replication-only solutions, our system works well both within larger servers and across clusters. Unlike virtualization only solutions, our system provides better performance and more flexibility in the deployment.