Each replica in a repository family is updated using a family operations stream enabling 100% consistency of all replicas after replication.
Network and server outages happen. MSP enables fully automatic catch-up of all write operations done at the sites that remained up. No administrator interaction required.
When MSP is deployed in a cluster it will enable a fully highly available environment.
When a remote disaster recovery site is configured all repositories replicated to it will be kept up to date with all of the other operational replicas.
You can choose which sites get which repositories. This enables access policies required in specific industries. It also enables lower cost (smaller size) servers for locations that require access to fewer repositories.
Repository “read” operations are done with minimal latency and without any interaction with MSP at all meaning full Subversion operational speed minimizing the time required.
Configurable data redundancy policies enable best performance for the compromise between redundancy, recoverability and speed.