GCP releases
Google Cloud releases and Terraform Google provider. New features, breaking changes, security advisories and deprecations - each summarised in plain English and updated continuously.
Tracking 330 GCP releases · Updated
- Google Cloud release notes datagcpgaengineer ·
BigQuery Remote Functions: Custom Paths in Endpoint URL
BigQuery remote functions now support custom paths in the endpoint URL, allowing reuse of a single Cloud Run service for multiple functions via path suffixes. This enhancement provides greater flexibility in managing remote function endpoints. The feature is now generally available.
feature - Google Cloud release notes datagcpengineer ·
AlloyDB for PostgreSQL: Configure read pool autoscaling cooldown
AlloyDB for PostgreSQL now allows configuration of a cooldown period before read pool instances autoscale. This feature provides more control over scaling events, potentially reducing costs and improving stability. It is available for users managing read pool instances and is detailed in the instance scaling documentation.
feature - Google Cloud release notes datagcpgaengineer ·
Firestore integrates with Google Cloud Knowledge Catalog
Firestore now supports searching and managing resources via Google Cloud Knowledge Catalog. This integration allows users to leverage Knowledge Catalog's metadata management capabilities for Firestore data. The feature is available now, enhancing data discovery and governance for Firestore users.
feature - Google Cloud release notes datagcpgaengineer ·
Firestore adds MongoDB compatibility and Knowledge Catalog integration
Google Cloud Firestore now offers MongoDB compatibility, allowing users to manage resources with familiar tools. This update integrates with Knowledge Catalog, enhancing metadata management and access for Firestore resources. This feature is now available for Firestore users seeking improved compatibility and data governance.
feature - Google Cloud release notes datagcpengineer ·
BigQuery Data Transfer Service adds Facebook Ads report support
The BigQuery Data Transfer Service has expanded its Facebook Ads connector to support additional reports, including AdInsightsMMM, Ads, and Campaigns. This enhancement allows for more comprehensive data ingestion from Facebook Ads into BigQuery. The update benefits users who rely on Facebook Ads data for analytics and reporting. The new reports are available now within the BigQuery Data Transfer Service.
patch - Google Cloud release notes datagovernancegcpengineer ·
GCP Knowledge Catalog Adds Data Lineage Search
The Data Lineage API in Google Cloud's Knowledge Catalog now supports the searchLineageStreaming method. This feature enables breadth-first searches for lineage links, making it easier for engineers and architects to trace data flow and dependencies. This enhancement is available now for users of the Knowledge Catalog.
feature - Google Cloud release notes dataawsgaengineer ·
Bigtable Enterprise Plus adds 365-day backup retention
Bigtable Enterprise Plus edition now supports configuring backup retention periods up to 365 days. This enhancement allows for longer data recovery options, benefiting organizations with extended compliance or historical data needs. The feature is now generally available for all users of the Enterprise Plus edition.
feature - Google Cloud release notes datagovernanceaws ·
Cloud Storage Lifecycle Management to Delete Empty Folders
Starting August 26, 2026, Cloud Storage will automatically delete empty folders in buckets with hierarchical namespace enabled when they meet Object Lifecycle Management conditions. This change impacts users managing object lifecycles who rely on empty folders remaining intact. The update aims to simplify data cleanup and reduce storage costs for hierarchical namespace buckets.
breaking - Google Cloud release notes datainfragcp ·
Oracle Database@Google Cloud expands to Australia and Italy
Oracle Database@Google Cloud has been expanded to new regions in Australia and Italy, providing customers with enhanced local access to Exadata Database Service, Base Database Service, and Goldengate. This expansion aims to improve performance and compliance for workloads running on Google Cloud infrastructure. The update is relevant to organizations utilizing Oracle databases within Google Cloud, particularly those requiring data residency in these new locations.
feature - Google Cloud release notes datagcpsnowflakegapreviewengineer ·
Looker 26.8 Enhancements: CI GA, Conversational Analytics, In-Database Models
Looker 26.8 introduces several new features and enhancements, including the general availability of Continuous Integration (CI) and expanded support for Conversational Analytics. Notably, it now integrates with in-database analytic models for BigQuery and Snowflake connections, enabling consistent semantic definitions across tools. Many new capabilities, such as granular dashboard sizing, enhanced observability, agentic workflows, and publishable conversational agents, are available in preview.
feature announcement - Google Cloud release notes datagcpengineer ·
BigQuery Simba ODBC Driver Updated
An updated version of the Simba ODBC driver for BigQuery has been released. This update likely includes performance improvements or bug fixes for connecting to BigQuery. Users who rely on the ODBC driver for their BigQuery integrations should consider updating to this new version.
patch - Google Cloud release notes datagcpengineer ·
Managed Service for Apache Spark: Image Rollback
Several subminor image versions for the Managed Service for Apache Spark have been rolled back due to issues. This change affects users who recently updated to these specific image versions. The rollback applies to Debian 12, Rocky 9, and Ubuntu 22 (including ARM variants) for version 2.2.82.
announcement - Google Cloud release notes datagcpgapreviewengineer ·
Google Cloud Knowledge Catalog: Data Products GA, New Features
Google Cloud's Knowledge Catalog has reached General Availability for its Data Products feature, enabling curated data asset packaging for business problems. Key enhancements include approval workflows for consumption, automated documentation generation with Gemini, and service account support. A preview feature allows programmatic interaction with data products via a remote Model Context Protocol server for data applications and AI agents.
feature - Google Cloud release notes datagcpgaengineer ·
Cloud Database Migration Service supports Cloud SQL for SQL Server sources
Cloud Database Migration Service now offers dedicated support for Cloud SQL for SQL Server as a source in homogeneous SQL Server migrations. This enhancement simplifies migrations by automatically exporting backups and uploading them to Cloud Storage. The feature is now generally available and is relevant for engineers and architects migrating SQL Server databases to Cloud SQL.
feature - Google Cloud release notes datagcpgaengineer ·
Oracle Database@Google Cloud adds GoldenGate support
Oracle Database@Google Cloud now supports Oracle Cloud Infrastructure GoldenGate, allowing users to create deployments and connections for data replication and transformation between systems. This new capability is now Generally Available. This enhancement is for Oracle Database users operating on Google Cloud.
feature - Google Cloud release notes dataaigcpgapreviewengineer ·
BigQuery Enhances Query Execution and Python UDFs
BigQuery can now re-execute queries to proactively detect performance or correctness regressions at no extra cost. Python User-Defined Functions (UDFs) are now generally available, allowing integration with PyPI libraries and external services. Additionally, a new AI.AGG function is available in preview for semantic aggregation of unstructured data using natural language.
feature announcement - Google Cloud release notes datagcpengineer ·
Managed Service for Apache Spark: Shuffle Partition Config Type Change
Google Cloud has updated the Managed Service for Apache Spark, formerly Dataproc on Compute Engine, by changing the configuration type for Spark shuffle partitions from integer to string for versions 2.2.82+ and 2.3.30+. This change requires users programmatically setting this configuration to update their code to use string literals for compatibility. Users setting the configuration via command-line, properties files, or Spark SQL commands are unaffected. This adjustment impacts specific subminor cluster image versions and aims to ensure consistent configuration handling.
breaking announcement - Google Cloud release notes datagcpengineer ·
Spanner graph query performance optimized with factorized execution
Spanner now supports factorized execution to optimize graph query performance by eliminating duplicate intermediate results. This feature benefits engineers and architects working with graph patterns in Spanner, potentially reducing query latency. The optimization is applied using the @{factorize_mode} hint at the query or pattern traversal level.
feature - Google Cloud release notes datagcpengineer ·
Bigtable enables row-affinity routing for standard app profiles
Google Cloud Bigtable now supports row-affinity routing for standard app profiles, which can improve read latency for geographically distributed applications. This feature is now available for enabling via the Google Cloud console. It primarily benefits engineers and architects managing globally distributed Bigtable deployments.
feature - Google Cloud release notes datagcpgapreviewengineer ·
BigQuery Enhances Reservation Management and Studio
BigQuery now allows grouping reservations to prioritize idle slot sharing, offering better control over high-priority workloads. A new organization policy feature, currently in preview, enables custom control over workload management resources. Additionally, BigQuery Studio now supports Git repositories for managing and versioning SQL scripts and notebooks, also in preview.
feature