Data Protection

Tungsten integrates advanced replication and backups to ensure complete protection from the DBA's worst nightmare: losing data. Tungsten makes it easy to set up live copies that protect from failures as well as administrative data deletion. Built-in consistency checks ensure data integrity so users can be confident that replicas are ready for use and correct.

Data Loss and Data Consistency

Automated consistency checks. Perform incremental checks of consistency between master and slave databases, without stopping the cluster and without causing a performance hit on applications.

Zero-data loss for planned and unplanned failovers. Tungsten has multiple options to eliminate corruption or loss of data when failing over from an old master to a promoted slave. These include flush events to ensure transactions clear from the master to the slave, specialized log consistency checks, and failure recovery procedures.

Recovery from admin data deletion. Tungsten can delay replication to offer a "window of opportunity" to recover data lost through administrative mistakes on the master, such as deleting application tables.

Integrated backup/restore. Tungsten integrates backups using file system snapshots and also through database dumps. Tungsten replication is tightly integrated with backup and restore. Users can dump or restore databases with a single command. In addition, Tungsten supports slave "auto-provisioning" in which newly configured replicators automatically load the most recent backup on start-up, a feature that is ideally suited for rapid slave provisioning. Slaves can even roll forward to a specific point in the log. Tungsten can integrate with a wide variety of backup software, so users can continue to use any backup with which they are familiar.

Disaster Recovery

Provision databases on remote site. Tungsten can provision slaves on a remote site using backups. This allows remote, live copies at a fixed point in time.Replicate data to remote site. Tungsten can similarly replicate to slaves on a remote site. This allows users to maintain fully-up-to-date live copies on the remote site.

Replicate database log to alternate host. The database log is the Achilles heel for both high availability and performance. This feature moves the current database log to to a slave machine, which means that database transactions get off the master host faster. It also has the potential to speed up replication by a factor of three or more.

Secure transport and data encryption. Secure transport with encryption protected replicated data as they move across publish
networks. It also seals up security holes that many replicator tools open up.