Highly Available, Resilient Data
Ensure uninterrupted, high performance queries during failure of a server, rack or region
Set availability goals at the database, table or row level
Avoid costly and complex failures and data loss
Eliminate planned and unplanned downtime
Maintain availability during failures
CockroachDB was architected to survive lost nodes without impacting data availability
- 3x data replication: When one replica is lost, remaining replicas continue serving reads and writes
- Self-healing infrastructure: data automatically rebalances when a node is lost and heals when you add back a node
Active-active availability
Avoid data loss
Eliminate the data loss, complexity, and cost that comes with an active-passive backup setup. In CockroachDB, writes occur concurrently across replicas of your data, so even if you lose a replica, you don’t lose data.
High availability docs
Customized survival
Set custom survival and availability goals
Different databases often have different availability requirements. With just a few declarative SQL statements, define survival goals for your databases so they handle availability zone or region loss without impact to performance.
Survival goals docs
Online rolling upgrades
Eliminate planned downtime
Never schedule a maintenance window again. CockroachDB’s live rolling upgrades feature lets you take down one node at a time for upgrades while the database continues serving reads and writes.
Upgrade overview
Distributed backup and restore
Distributed backup and restore
Ensure you’re ready for worst-case-scenario with backup and restore functionality that scales with your data, distributed across all nodes in your cluster. Get full and incremental backups, scheduled backups, and point in time recovery.
Disaster recovery overview
Backup docs
