Apigee Advanced API Security
Announcement
On March 17, 2026 we released an updated version of Advanced API Security abuse detection
Feature
VPC-SC support in abuse detection
This release includes full support in Advanced API Security abuse detection for VPC-SC customers. This includes support for VPC-SC with the Advanced Anomaly Detection ML model used for abuse detection, as well as detection exclusion lists.
For usage information, see Abuse detection in the documentation.
Apigee X
Announcement
On March 17th, 2026, we released an updated version of Apigee (1-17-0-apigee-5).
Fixed
| Bug ID | Description |
|---|---|
| N/A | Updates to infrastructure and libraries. |
App Engine flexible environment Ruby
Feature
Support for Ruby 4.0 runtime is in General Availability.
App Engine standard environment Ruby
Feature
Support for Ruby 4.0 runtime is in General Availability.
BigQuery
Feature
In BigQuery ML, you can now automatically deploy open models to Vertex AI endpoints. Automatically deployed models offer the following benefits:
- Automatic Vertex AI resource management
- Reserve open model resources by using Compute Engine reservations
- Automatic or immediate open model undeployment to save costs
This feature is generally available (GA).
Cloud Data Fusion
Fixed
Cloud Data Fusion version 6.11.1.2 is generally available (GA).
This release includes the following changes:
Fixed the triggers panel in the pipeline details page to display the correct triggers count on initial load (CDAP-21230).
Updated the GraphQL query mechanism to cache the pipelines list and fix the long loading screen in the deployed pipelines list page (CDAP-21229).
Cloud Run
Feature
Support for Ruby 4.0 runtime is in General Availability.
Cloud Run functions
Feature
Support for Ruby 4.0 runtime is in General Availability.
Cloud SQL for PostgreSQL
Feature
You can now cancel an in-place major version upgrade operation during the main upgrade phase, when the upgrade is actually being performed.
For more information, see Cancel the major version upgrade.
Google SecOps
Announcement
Unified Feature Role-based Access Control (RBAC) is now in General Availability (GA). This enables administrators to manage feature access control for Google SecOps including SOAR by leveraging Google Cloud IAM instead of managing it separately for SIEM and SOAR.
You can enable it by migrating the legacy SOAR permission groups and permissions to Google Cloud IAM through a self-service migration available from January 26, 2026. Please check the documentation and video for full instructions.
This update is available to all customers who have completed Stage 1 of the SOAR migration to Google Cloud.
Google SecOps SOAR
Announcement
SOAR Permission Groups migration to Google Cloud IAM is now in General Availability (GA). You can now leverage Google Cloud IAM for precise, granular feature access, moving away from legacy permission groups.
You can enable it by migrating the legacy SOAR permission groups and permissions to Google Cloud IAM through a self-service migration available from January 26, 2026. Please check the documentation and video for full instructions.
This update is available to all customers who have completed Stage 1 of the SOAR migration to Google Cloud.
Looker
Feature
The Visualization Assistant, which lets you customize formatting options for Looker visualizations in natural language with Gemini assistance, is now generally available. This assistant can be enabled by turning on the Gemini in Looker and Looker Assistants settings on the Gemini in Looker page in the Platform section of the Admin panel.
Feature
Now available in preview, Gemini in Looker can automatically generate Quick Start analyses for Explores. This feature can be enabled by turning on the AI-assisted Quick Starts setting on the Gemini in Looker page in the Platform section of the Admin panel.
Feature
Now available in preview, you can use Gemini in Looker to assist you in writing Looker expressions for table calculations and custom fields. This feature can be enabled by turning on the Expression Assistant setting on the Gemini in Looker page in the Platform section of the Admin panel.
Feature
Looker Connections Settings page has a new option, Disable Connection, that allows a Looker admin to disable a connection in cases where there are downstream issues with the database, instead of killing queries manually or allowing queries to remain in the query queue. When the connection is disabled, Looker will not send queries to the database and will return an error message to users.
Feature
The Self-service Explores feature is now generally available.
In addition, if your Looker admin has enabled your Looker instance to support OAuth for Google Sheets uploads, the self-service Explores feature supports uploading data from Google Sheets using Google Drive navigation.
Announcement
The following features will begin rolling out on March 17, 2026.
Feature
The content certification feature is now generally available. In Looker 26.4, the following support has been added:
- LookML Explores can now be certified.
- Looker admins can now automatically certify all current and future LookML dashboards and LookML Explores.
- When Enhanced Search is enabled, searching for content now includes the ability to sort and filter search results based on certification status.
Feature
The Enhanced Search preview feature is now available.
When enabled, Enhanced Search helps you find saved content on your instance using Gemini in Looker. Enhanced Search moves beyond keyword matching to interpreting the conceptual meaning of your search queries, letting you search for saved content using business terms or analytical questions (for example, “total customer acquisition cost”).
You can filter your search results to limit them to specific content types, folders, content creators, creation dates, modification dates, and the content certification status.
To enable Gemini in Looker assistance with searching content, you must also enable the Semantic Search setting on the Gemini in Looker page in the Platform section of the Admin panel.
Feature
Available in preview for Looker connections to Snowflake or Google BigQuery databases, Looker supports custom calendars, such as fiscal or retail calendars. The custom calendar feature lets your Looker developers create a LookML model of a custom calendar table in your database and then create custom calendar dimension groups that use the custom calendar. Your end users can then create Explore queries using the custom calendar dimension timeframes.
In addition, your LookML developers can create period-over-period measures based on custom calendar dimension groups.
Feature
The Increased Row Limit preview feature feature is now available, and is disabled by default.
When enabled, the Increased Row Limit preview feature lets admins set row limits up to 50,000 rows or datapoints for map charts, scatterplot charts, and table charts. Admins can set the limits for each visualization type in the Visualization limits setting on the Content guardrails admin page.
Table charts, scatterplot charts, and Google Maps charts that have increased row limits are subject to additional limitations when you download, send, or schedule dashboards in PDF format. See Downloading or delivering dashboards in rendered formats for more information about scheduling and downloading dashboard PDFs with increased row limits.
Note: This item was originally published on March 16, 2026 and was updated on March 17, 2026.
Feature
The Gemini in Looker assistant that helps you in generate LookML parameters is now available to enable individually from other Gemini in Looker features. This assistant, which continues to be available in preview, suggests LookML parameters that are based on the natural language prompts that you add to your project files. This assistant can be enabled by turning on the Gemini in Looker and LookML Assistant settings on the Gemini in Looker page in the Platform section of the Admin panel.
Feature
The Continuous Integration (CI) feature is updated for better integration with the Looker user interface:
- Your Looker instance will be migrated to the new version of Looker CI. The migration will take a few minutes.
- While your instance is migrating, any existing CI suites will not appear on the CI Suites page, and the history on the CI Runs page will be empty. During this time, do not create new suites or initiate manual runs.
- When the migration is complete, the Suites and Runs pages will be repopulated.
- If you encounter an issue during the migration, please report it using the Looker} Continuous Integration – Preview Issue Tracker form.
The updated Looker Continuous Integration feature includes the following:
- Your Looker admin no longer needs to manually add CI users. After your Looker instance is enabled for CI (or once your instance is updated to the new CI, if your instance was already enabled for CI), Looker CI users are created automatically. Looker CI users are created in the Looker CI Users user group with the Looker CI Users role. A Looker admin can view the Looker CI users from the Service Accounts tab of the Users Admin page.
- A new GitHub table on the Continuous Integration Admin page lists the GitHub repositories that have been granted access to the Looker CI GitHub application, which enables Looker CI to be triggered automatically when pull requests are created.
Feature
Conversational Analytics now offers two modes for asking questions. Fast mode allows you to get answers more quickly. Thinking mode allows you to ask more complex questions and test your agent’s capabilities.
Memorystore for Valkey
Feature
You can now deploy instances in the asia-southeast3 (Bangkok), europe-north2
(Stockholm), and northamerica-south1 (Mexico)
regions.
Source: Google Cloud Platform
