AI Hypercomputer
Issue
Workloads on A4 VMs might experience interruptions due to a firmware issue for NVIDIA B200 GPUs. To help prevent the issue, we recommend resetting the GPUs on A4 VMs at least once every 60 days. For more information, see the known issue.
Bigtable
Feature
In the Google Cloud console, you can import data into Bigtable using the Dataflow job builder, a web interface for building and running Dataflow pipelines. This integration provides a pre-populated template to import data from Pub/Sub to Bigtable. You can also create a custom job to import data from other sources. For more information, see Import and export data.
Cloud Composer
Deprecated
The following Cloud Composer versions and builds have reached their end of support period: composer-3-airflow-2.9.3-build.11, composer-2.10.1-*.
Changed
New Airflow builds are available in Cloud Composer 3:
- composer-3-airflow-3.1.0-build.6
- composer-3-airflow-2.10.5-build.23 (default)
- composer-3-airflow-2.9.3-build.43
Fixed
Fixed an issue where the Copy button on the DAG details page in the Airflow UI was copying incorrect content.
Changed
New images are available in Cloud Composer 2:
Feature
Database retention policy is now available in environments with Airflow 3, starting with composer-3-airflow-3.1.0-build.5.
This change is gradually rolled out in several releases and is available in the following regions in this release: africa-south1, asia-east1, asia-northeast2, asia-northeast3, asia-south2, asia-southeast2, asia-southeast3, australia-southeast2, europe-central2, europe-north2, europe-southwest1, europe-west10, europe-west12, europe-west15, europe-west4, europe-west6, europe-west8, europe-west9, me-central1, me-central2, me-west1, northamerica-northeast1, northamerica-northeast2, northamerica-south1, southamerica-east1, southamerica-west1, us-east5, us-south1, us-west3, us-west4.
Cloud Logging
Changed
The default setting for the time-range selector for the Logs Explorer is now five minutes. The previous default was one hour.
Compute Engine
Feature
The memory-optimized X4 machine series offers additional bare metal machine types with 6 TB, 8 TB, and 12 TB of memory. For more information, see X4 machine series.
Issue
Workloads on A4 VMs might experience interruptions due to a firmware issue for NVIDIA B200 GPUs. To help prevent the issue, we recommend resetting the GPUs on A4 VMs at least once every 60 days. For more information, see the known issue.
Gemini
Issue
Known issue for the model selection feature in VS Code Gemini Code Assist
The VS Code Gemini Code Assist model selection feature incorrectly appears for users of the free tier of Gemini Code Assist for individuals.
- The feature shouldn’t appear for such users and does not function for them.
Issue
Known issue for the model selection feature in VS Code Gemini Code Assist
The VS Code Gemini Code Assist model selection feature incorrectly appears for users of the free tier of Gemini Code Assist for individuals.
- The feature shouldn’t appear for such users and does not function for them.
Gemini Enterprise
Feature
Unified data Source creation and actions flow
The new unified process for creating a data store enables you to create the data store and the associated actions in a single, streamlined flow. This functionality allows Gemini Enterprise to perform tasks directly within the connected application.
The new Actions page shows you all the enabled actions for each data store, which are currently available in public preview. Actions are currently enabled for Microsoft SharePoint, Microsoft Outlook, Microsoft OneDrive, Jira Cloud, Dropbox, and ServiceNow, and are available in public preview.
This launch includes support for the following data stores:
General Availability (GA)
- Microsoft SharePoint
- Microsoft Outlook
- Microsoft OneDrive
- Google Drive
- Google Calendar
- Gmail
Public Preview
- Jira Cloud
- Confluence Cloud
- Box
- Microsoft Teams
- Dropbox
- ServiceNow
Generative AI on Vertex AI
Feature
Gemini 2.5 Flash with Gemini Live API Native Audio
Gemini 2.5 Flash with Gemini Live API Native Audio (gemini-live-2.5-flash-native-audio) is Generally Available (GA).
This model features cutting-edge native audio functionality for
Gemini Live API, including enhanced voice quality and adaptability, Proactive Audio, and Affective Dialog.
Google Kubernetes Engine
Feature
GKE Autopilot now supports N4A machine types in Public Preview, available on clusters running version 1.34.1-gke.3403001 or later.
SAP on Google Cloud
Announcement
New SAP certification: 6, 8, and 12 TB X4 machine types for SAP HANA scale-up workloads
For use with SAP HANA scale-up (OLTP and OLAP) workloads running on Google Cloud, SAP has certified the following Compute Engine X4 machine types:
x4-480-6t-metal– 6 TB memoryx4-480-8t-metal– 8 TB memoryx4-960-12t-metal– 12 TB memory
With this certification, the names of all X4 machine types have been updated to
convey the memory they offer. For example, x4-megamem-960-metal is renamed to
x4-960-16t-metal because it offers 16 TB memory. This change is backwards
compatible.
For more information, see X4 memory-optimized bare metal machine types.
Security Command Center
Feature
AI Protection is generally available (GA) in the Security Command Center Enterprise tier and is available as a Preview in the Security Command Center Premium tier.
Changed
Multiple pages in Security Command Center Standard have been improved:
- The Risk overview page is enhanced to provide separate views for misconfiguration, vulnerabilities and identity-related findings.
- The Findings page includes predefined filter views for vulnerabilities and identity findings.
- The left navigation has been updated.
Text-to-Speech
Feature
Chirp 3: HD voices now support speech synthesis in Preview for the Punjabi (pa-IN) language. For more information, see Chirp 3: Language Availability.
Source: Google Cloud Platform
