Examining Versionless Apache Spark™: AI-powered upgrades and seamless stability for 2 billion workloads | Databricks Blog
Databricks removed the pain of Spark version upgrades with the first fully versionless Spark engine for Serverless notebooks and jobs. Versionless Spark exposes stable client APIs and automatic version upgrades, supported by AI-powered release management - resulting in a 99.99999% upgrade success rate. Versionless Spark ensures workload stability and frictionless upgrades so that workloads always run on the latest version, benefiting from newly released performance and security improvements with zero code changes.
