Data Filtering

Tungsten offers powerful filtering capabilities that can change or drop replicated SQL both on the master as well as slaves.

Extensible data filters. Tungsten includes built-in filters, and users can easily develop new ones using either Java or Javascript. Filtering is so useful that Tungsten itself uses filters to implement time-delay replication, handle flush operations, and assign events to channels for parallel replication.

Data filter libraries. Tungsten integrated filtering technology can change SQL "on the fly" from the source dialect to the target. This enables a much wider range of SQL to be replicated across database platforms such as Oracle to MySQL.  Scaling applications without changes is the holy grail of all database clusters. Tungsten SQL Routing helps master/slave clusters appear as a single database to your applications, which is the key feature required to reduce application impact.