Start of main content
Talk type: Talk
Asynchronous replication is a well-known technique to provide an independent replica of a database for different purposes including disaster recovery, geo-replication or scaling read workload. Traditionally, a log of changes from source replica is treated as an input for target replica database. If source and target replicas are distributed databases that handle high workload, putting all updates to a single log of changes could be nearly impossible. For instance, YDB is a Distributed SQL Database that is built out of reliable components called tablets, each tablet has its own recovery log, and user update may modify data in different tablets in one transaction. Multi tablet transactions provide ACID guarantees via an internal mechanism of distributed transaction execution.
So, there is no single log of changes that could be transferred to the target database. In this talk we present an approach to asynchronous replication in YDB that provides the following characteristics: