Google Releases Spanner Omni for On-Premises Deployments

Google has announced the general availability of Spanner Omni, a version of its distributed SQL database designed to run on customer-managed infrastructure, including on-premises data centers, Google Cloud, Amazon Web Services, and local laptops. To untether the database from its native cloud environment, Google replaced its Colossus distributed file system with a local abstraction layer and swapped its TrueTime atomic clock and GPS hardware for a software-based time synchronization system. Despite these changes, Google's internal benchmarks show the system can still process millions of queries per second across petabytes of data in a single regional deployment.

The database retains core features like Paxos consensus, automatic sharding, and synchronous replication. It supports GoogleSQL, PostgreSQL, and Spanner Graph Language, combining relational, graph, key-value, and vector models with a columnar engine. The general availability release introduces TLS encryption, audit logging, backup and restore capabilities, and stateless worker nodes to offload background tasks. Additionally, Model Context Protocol support via the MCP Toolbox allows AI agents to read database schemas for operational memory.

However, running Spanner Omni shifts operational responsibilities to the customer. Google does not offer an availability service-level agreement (SLA) for Omni, meaning users must manage their own failure domains, backups, and upgrades. Monitoring is handled via Prometheus and Grafana rather than Google Cloud Monitoring. Licensing is split into two tiers: a free Developer Edition, which is restricted to non-production use and limits permanent deployments to single servers with 4 vCPUs or fewer, and a Commercial Edition sold as an annual subscription based on vCPU usage with no public pricing.