Google Cloud Platform Newsletter

Check Archive for older issues

I apologize in advance for longer issue than usual, but last week was quite fruitful with many announcents from Cloud Next.

News

229 things we announced at Google Cloud Next '25 – a recap - Everything important in one place.

Welcome to Next ‘25 - Recent advances on Google Cloud.

Day 1 at Google Cloud Next 25 recap - Google Cloud Next 25 kicked off in Las Vegas, focusing on how AI can transform work and improve lives.

Next 25 developer keynote: From prompt, to agent, to work, to fun - The keynote introduced new capabilities such as Gemini 2.5 Pro, Agent Development Kit (ADK), Vertex AI Agent Engine, and Gemini Code Assist Kanban board. It also highlighted how Gemini can be used in various domains, including baseball pitch analysis and data forecasting.

High performance storage innovations for your AI workloads - Google Cloud introduces high-performance storage innovations to optimize AI workloads. Rapid Storage offers sub-millisecond latency and high throughput, while Anywhere Cache improves read-storage latency by up to 70%. Google Cloud Managed Lustre provides a fully managed parallel file system with sub-millisecond latency and high throughput. Storage Intelligence analyzes object metadata to generate storage insights and optimize costs.

Chrome Expands AI-Powered Enterprise Search and Enterprise Browser Protections - Chrome Enterprise is expanding its AI-powered enterprise search capabilities by integrating with Google Agentspace, allowing employees to find information from various sources directly within Chrome.

Delivering an application-centric, AI-powered cloud for developers and operators - Google Cloud introduces an application-centric, AI-powered cloud experience for developers and operators. New features in Gemini Code Assist and Gemini Cloud Assist provide AI assistance throughout the application lifecycle, from design and development to deployment and management.

Driving secure innovation with AI and Google Unified Security - Google announced Google Unified Security, a converged security solution powered by AI that brings together best-in-class security products for threat intelligence, security operations, cloud security, and secure enterprise browsing, along with Mandiant expertise.

Transforming customer experiences with AI agents and the next generation Customer Engagement Suite - Google Cloud's Customer Engagement Suite transforms customer experiences with AI agents. The next generation of Conversational Agents simplifies building AI agents, enables highly engaging customer experiences, and automates work across operations. Companies like Verizon, Wendy's, and Mercedes-Benz are already using the suite to enhance customer service and loyalty.

New services to help enterprises adopt AI from Google Cloud Consulting - Google Cloud Consulting offers new services to help enterprises adopt AI, including Agentspace Accelerator, Optimize with TPUs, and Oracle on Google Cloud. Delivery Navigator, an AI-enhanced platform, provides resources and best practices for cloud-based projects.

Articles, Tutorials

Infrastructure, Networking, Security, Kubernetes

Driving enterprise transformation with new compute innovations and offerings - Google Cloud announces new compute innovations and offerings, including C4D VMs with improved performance, C4 VMs with new capabilities, H4D VMs for HPC workloads, M4 VMs for SAP workloads, Z3 for storage-intensive workloads, Nutanix Cloud Clusters on Google Cloud, and more options for VMware Engine. The company also introduces next-generation block storage with Hyperdisk, including Hyperdisk Storage Pools and Hyperdisk Exapools, and provides greater efficiency and control with managed instance groups.

Introducing Ironwood TPUs and new innovations in AI Hypercomputer - Google Cloud introduces new innovations in AI Hypercomputer, including Ironwood TPUs, enhanced networking, and software capabilities for training and inference. Ironwood TPUs offer 5x more peak compute capacity and 6x the high-bandwidth memory capacity compared to the previous generation.

H4D VMs: Next-generation HPC-optimized VMs - Google Cloud has introduced H4D VMs, a new machine type optimized for high-performance computing (HPC). Powered by 5th Generation AMD EPYCTM Processors, H4D VMs offer improved whole-node VM performance and memory bandwidth. They also feature Cloud RDMA on Titanium, enabling low-latency and 200 Gbps network bandwidth. H4D VMs significantly outperform previous generations of AMD-based VMs, delivering faster simulations and analysis across various HPC applications and benchmarks.

Colossus: the secret ingredient in Rapid Storage’s high performance - Rapid Storage, a new object storage service from Google Cloud, leverages the stateful protocol of Colossus, Google's internal zonal cluster-level file system, to provide sub-millisecond read/write latency and durable appends. With its gRPC-based streaming protocol, Rapid Storage supports 20 million requests per second in a single bucket, making it ideal for AI/ML data preparation, training, and checkpointing, distributed database architecture optimization, batch and streaming analytics processing, video live-streaming and transcoding, and logging and monitoring.

Enabling global scientific discovery and innovation on Google Cloud - Google Cloud is introducing new capabilities to enable scientific discovery and innovation. Researchers can now deploy and use supercomputing clusters powered by the latest H4D VMs powered by AMD CPUs, and A4/A4X VMs powered by the latest NVIDIA GPUs.

Kubernetes, your AI superpower: How Google Kubernetes Engine powers AI innovation - Google Kubernetes Engine (GKE) offers significant improvements to help platform teams succeed with AI. Cluster Director for GKE enables the deployment and management of large clusters of accelerated VMs with compute, storage, and networking operating as a single unit. GKE Inference Quickstart simplifies infrastructure selection and AI model deployment with benchmarked performance characteristics. GKE Inference Gateway provides intelligent routing and load balancing for AI inference on GKE. A new container-optimized compute platform is rolling out on GKE Autopilot, optimizing cloud usage and cost savings. Gemini Cloud Assist Investigations helps with GKE troubleshooting, reducing the time to understand root causes and resolve issues. RayTurbo on GKE, launching later this year, delivers superior GPU/TPU performance, rapid cluster startup, and robust autoscaling.

New GKE inference capabilities reduce costs, tail latency and increase throughput - Google Cloud Next introduced new capabilities for GKE inference, including GKE Inference Quickstart, GKE TPU serving stack, and GKE Inference Gateway. These capabilities help reduce serving costs by over 30%, tail latency by 60%, and increase throughput by up to 40% compared to other managed and open-source Kubernetes offerings. Customers like HubX and BentoML have already seen significant benefits from using these new capabilities.

What’s new with Google Cloud networking - Google Cloud Next 2025 introduced several innovations in cloud networking to support the demands of the AI era. These include AI-optimized networking for massive data ingestion, unprecedented cluster scale, zero-trust RDMA security, and accelerated GPU-to-GPU communication.

Cloud WAN: Connect your global enterprise with a network built for the AI era - Google Cloud WAN is a fully managed, reliable, and secure enterprise backbone to transform enterprise wide area network (WAN) architectures. It leverages Google’s planet-scale network, which is optimized for application performance. Cloud WAN provides up to 40% faster performance compared to the public internet, and up to a 40% savings in total cost of ownership (TCO) over a customer-managed WAN solution. Cloud WAN offers two key use cases: providing high-performance connectivity between geographically dispersed data centers, and connecting branch and campus environments over our Premium Tier network.

Google's AI-powered next-generation global network: Built for the Gemini era - Google's AI-powered global network, built for the Gemini era, addresses the challenges of the AI era with four new design principles: exponential scalability, beyond-9s reliability, intent-driven programmability, and autonomous network. It features advancements such as multi-shard network architecture, multi-shard isolation, region isolation, protective reroute, fully intent-driven programmability, and autonomous network operations powered by machine learning. This network offers cloud customers the capacity, elasticity, and scale to deploy and leverage AI effectively, while end users benefit from expanded global reach, resilient mission-critical applications, zero-trust security, and performant network for power-intensive real-time apps.

Bringing Gemini and Google Agentspace to you on-premises - Google Distributed Cloud (GDC) now offers Gemini, Google's most capable AI models, on-premises, with a public preview starting in Q3 2025.

Simplifying AI Development at scale: Google Cloud Integrates Anyscale's RayTurbo with GKE - Google Cloud and Anyscale have partnered to integrate Anyscale RayTurbo, a high-performance runtime for Ray, with Google Kubernetes Engine (GKE).

Windows Remote Desktop Protocol: Remote to Rogue - This blog post discusses a novel phishing campaign targeting European government and military organizations attributed to a suspected Russia-nexus espionage actor tracked as UNC.

App Development, Serverless, Databases, DevOps

AlloyDB AI drives innovation for application developers - AlloyDB AI, a cloud-based database service, introduces new capabilities for app developers. These include advanced semantic search with high-performance filtered vector search, automatic vector index maintenance, and improved search quality using the AlloyDB AI query engine and Vertex AI Ranking API. Natural language capabilities enable users and agents to gain insights from natural language questions. AlloyDB AI query engine unlocks deep semantic insights from enterprise data through AI-powered SQL operators. These innovations position AlloyDB as the foundation for agentic AI, evolving the database beyond data storage and conventional SQL querying.

Accelerate your analytics: New Bigtable SQL capabilities drive real-time insights - Bigtable SQL, now generally available, simplifies working with Bigtable's flexibility and speed using familiar SQL syntax. Continuous materialized views, in preview, offer real-time aggregation and analysis of data streams without the complexities of traditional materialized views. Ecosystem integrations, such as the open-source Apache Kafka Bigtable Sink and Apache Flink Connector for Bigtable, simplify building end-to-end applications. Bigtable CQL Client, also in preview, enables developers to migrate Cassandra applications to Bigtable with minimal code changes.

Supercharge SQL Server to PostgreSQL migrations with AI-powered code conversion in Database Migration Service - Google Cloud's Database Migration Service (DMS) now supports SQL Server to PostgreSQL migrations, enabling organizations to unlock the potential of open-source databases in the cloud.

Google Cloud and Oracle accelerate enterprise modernization with new offerings, regions, and capabilities - Google Cloud and Oracle are expanding their partnership to accelerate enterprise modernization. New offerings include Oracle Base Database Service, general availability of Oracle Exadata X11M, and additional enterprise-ready capabilities like customer managed encryption keys (CMEK).

Google Cloud databases supercharge the AI developer experience - Google Cloud is introducing new database capabilities, including generative AI capabilities in AlloyDB and agentic programming with MCP, MongoDB compatibility in Firestore, expanded Oracle services, and a SQL Server modernization solution.

Google Axion processors boost AlloyDB, Cloud SQL, major customers, and ISVs - Google Axion processors, custom Arm-based CPUs, are now available for Cloud SQL and AlloyDB for PostgreSQL managed databases, providing significant price-performance advantages for database workloads.

Introducing Firebase Studio and agentic developer tools to build with Gemini - Firebase Studio is a cloud-based, agentic development environment powered by Gemini that includes everything developers need to create and publish production-quality AI apps quickly.

Announcing Firestore with MongoDB compatibility, an enterprise-ready document database - Google Cloud announced Firestore with MongoDB compatibility, a serverless document database service that provides developers with an additional choice for demanding document database workloads. Firestore with MongoDB compatibility offers the agility of the popular MongoDB API and query language, combined with Firestore's scalable and highly available architecture.

GCP API Gateway for Cloud Run authenticated backend with Firebase Auth JWT Tokens - An exercise in finding the right Authorization Header to Authenticate Users.

Big Data, Analytics, ML&AI

What's new for Data Analytics in the AI driven, autonomous and agentic era - Google Cloud's Data & AI Cloud platform introduces several innovations to empower businesses with real-time data activation. Specialized agents for data engineers, data scientists, and analysts simplify tasks and boost productivity.

What's new with BigQuery — the autonomous data-to-AI platform - BigQuery, the autonomous data-to-AI platform, enables organizations to leverage enterprise data for AI agents, intelligent search, and AI-powered analytics. It offers a self-managing multimodal data foundation, advanced AI assistance, simplified governance, and enhanced enterprise capabilities. With BigQuery, businesses can accelerate their journey to AI-driven insights at a lower cost.

Looker adds AI-fueled visual, conversational data exploration, continuous integration - Looker, a business intelligence and data analytics platform, introduces AI-powered features and a new reporting experience. Users can now leverage conversational analytics to analyze data using natural language and Google's Gemini models. Looker reports enhance data storytelling and streamline exploration, with drag-and-drop functionality, granular design controls, and real-time collaboration. Continuous integration capabilities automate testing and validation of SQL and LookML changes, ensuring faster and more reliable development cycles.

Introducing BigQuery unified governance: universal, intelligent, and open - Google Cloud announced BigQuery unified governance, a set of powerful data governance capabilities that help enterprises manage and secure their data and AI assets. Key features include full-catalog search with semantic understanding, automated metadata curation, AI-powered knowledge engine, data products, business glossary, automatic at-scale cataloging of BigLake and object tables, automated anomaly detection, BigQuery sharing integration with Google Cloud Marketplace, stream sharing in BigQuery, stored procedure sharing in BigQuery, query template sharing in BigQuery, data policies on columns, and subquery support with row-level security.

Simplify your data platform migration with AI-powered BigQuery Migration Services - Google Cloud’s BigQuery Migration Services now offers several new innovations, including coverage for data science and expanded support for data engineering and data analytics workloads. These new capabilities span across four stages of a data platform migration: automated assessment and planning, automatic code translation, data migration, and validation. Customers can now benefit from automated discovery and assessment with estimated total cost of ownership, automated code translations, data, metadata, and permissions migration, and intelligent end-to-end validation.

34 gen AI success stories with customers and ISVs - In this blog post, we present 34 real-world examples of how companies are using generative AI to transform their businesses. These use cases span a wide range of industries, from financial services to healthcare to retail, and demonstrate the power of AI to automate tasks, improve decision-making, and create new products and services.

Scale enterprise search and agent adoption with Google Agentspace - Google Agentspace is a product that provides employees with access to the latest Google foundation models, powerful agents, and actionable enterprise knowledge. With Agentspace, customers can easily discover and adopt agents from the Agent Gallery, create custom agents with the no-code Agent Designer, and deploy Google-built agents such as the Deep Research and Idea Generation agents.

Building the industry’s best agentic AI ecosystem with partners - Google Cloud is committed to building the industry's best agentic AI ecosystem with partners. At Google Cloud Next, the company announced updates that will help partners build AI agents, power them with enterprise data and a choice of models, and bring them to customers through a new AI Agent Marketplace.

Gemini 2.5 brings enhanced reasoning to enterprise use cases - Gemini 2.5, Google's latest AI model, offers enhanced reasoning capabilities for enterprise use cases. Gemini 2.5 Pro excels in complex tasks requiring deep reasoning and coding expertise, while Gemini 2.5 Flash prioritizes speed, low latency, and cost-efficiency for high-volume applications. Vertex AI Model Optimizer helps choose the best model for specific needs, and the Live API enables real-time interactions with streaming audio, video, and text.

Vertex AI offers new ways to build and manage multi-agent systems - Vertex AI offers new ways to build and manage multi-agent systems. It combines an open approach with comprehensive platform capabilities to ensure agents perform reliably. Enhancements include the Agent Development Kit (ADK) for building agents with an open approach and deploying them with enterprise-grade controls, the Agent Engine for deploying AI agents to production with built-in testing, release, and reliability, and the Agent2Agent protocol for connecting agents across different ecosystems.

Vertex AI is now the only platform with generative media models across video, image, speech, and music - Google Cloud's Vertex AI platform now offers Lyria, a text-to-music model, making it the only platform with generative media models across video, image, speech, and music.

Power up your BigQuery analysis with Google's new geospatial datasets - Google Cloud has introduced new geospatial analytics datasets and capabilities from Earth Engine and Google Maps Platform, integrated directly into BigQuery. These capabilities enable users to access fresh, comprehensive, and global geospatial data within the familiar BigQuery platform, simplifying data discovery, integration, and analysis.

Building a RAG System for Heavy Machinery Engine Manuals - Developing a Retrieval Augmented Generation system to provide technicians with direct, unmodified instructions from engine manuals.

Much simplified function calling in Gemini 2.X models - In the latest Gemini 2.X models and the unified Google Gen AI SDK, function calling has been significantly simplified. Now, you can declare your functions with comments and pass them directly to the model, eliminating the need for special function declarations, API calls, and manual response handling.

Agentspace Connectors and Actions: Secure Integration Explained - Google Cloud Agentspace securely integrates with various data sources using its extensive connector library. It respects the source applications' access control lists (ACLs), ensuring that users only see data they have access to. Predefined Actions enable powerful interactions back into those systems, such as sending emails, creating meetings, or raising requests.

Building AI Agents with Google’s Agent Development Kit (ADK) as MCP Client — A Deep Dive (Full Code) - This article explores how to use the open-source Agent Development Kit (ADK) to build an agent setup assistant using the Model Context Protocol (MCP) with Google Gemini LLM as an MCP Client.

Conversing with GKE Clusters: Google’s Agent Development Kit in Action - Implementation of a GKE chatbot using ADK, showcasing its capabilities in interacting with Kubernetes clusters via natural language. The chatbot can list namespaces, pods, and deployments, and chain tools intelligently without hardcoded workflows.

Various

New Google Workspace cost-saving offer available for U.S. federal government - Google Workspace, a FedRAMP High authorized communication and collaboration platform, is now available at a significant discount for U.S. federal government agencies.

Global startups are building the future of AI on Google Cloud - This year, over 60% of the world's generative AI startups are building on Google Cloud, including Anthropic, Cohere, Magic, Mistral, and AI21 Labs. Google Cloud is also announcing new partnerships with accelerators and VCs, as well as new resources to help startups build and go to market more quickly.

Presenting our 2025 Google Cloud Partners of the Year - The 2025 Google Cloud Partners of the Year award winners have been announced, recognizing dozens of partners for their dedication, innovation, and collaboration in delivering impactful solutions on Google Cloud.

Slides, Videos, Audio

GCP Bytes Podcast - #14 In this episode we discuss; Tariffs, Google Next Announcements, NBN Baseline, GDG, Mantel Group APAC Partner of the year, VMWARE Ups Fees, Apple Appeal, BQ Repos, Virtual Repos, Oracle Breach, Australian Super Hack, Gemini 2.5 Pro, Vibe Coding.

Security Podcast - #218 IAM in the Cloud & AI Era: Navigating Evolution, Challenges, and the Rise of ITDR/ISPM.

Releases

Agent Assist - Agent Assist offers a UI Connector with Salesforce to integrate with voice conversations.

AlloyDB - AlloyDB for PostgreSQL supports a 1 virtual central processing unit (vCPU) configuration with 8GB of memory, which is suitable for development and sandbox environments. AlloyDB supports AI-assisted troubleshooting that helps you resolve complex database performance issues like slow queries and high load. AlloyDB for PostgreSQL supports parameterized secure views, which provide a secure interface for application developers by improving data security and row access control while using SQL. AlloyDB AI natural language (Preview) delivers secure and accurate responses for application end user natural language questions. AlloyDB AI query engine that builds on model endpoint management, and adds support for AI operators and Vertex AI multimodal and ranking models is available in (Preview). The alloydb_scann extension is updated to include the following vector search improvements. The alloydb_scann extension is updated to include the following vector search improvements in (Preview): You can enable auto-maintenance for your ScaNN index and let incrementally manage the index such that when your dataset grows, AlloyDB splits large outlier partitions, and tries to provide better QPS and search results. AlloyDB supports C4A Arm VMs on Google's custom-built Axiom processors. AlloyDB now supports managed connection pooling in Preview.

App Hub - You can create and manage your App Hub applications using app-enabled folders, now available in Preview. Gemini Cloud Assist in App Hub is supported in Preview.

Cloud Architecture Center - (New guide) Harness CI/CD pipeline for RAG applications: Shows how to implement a continuous integration (CI) and continuous deployment (CD) pipeline for a retrieval-augmented generation (RAG) application in Google Cloud. Design an optimal storage strategy for your cloud workload: Added guidance about Google Cloud Managed Lustre. Design storage for AI and ML workloads in Google Cloud: Updated to include Cloud Storage FUSE, Anywhere Cache, Hyperdisk ML, and Google Cloud Managed Lustre. (New guide) Optimize AI and ML workloads with Cloud Storage FUSE: Learn how to optimize performance for AI and ML workloads on Google Kubernetes Engine (GKE) by using Cloud Storage FUSE. (New guide) Oracle E-Business Suite with Oracle Exadata in Google Cloud: Shows how to build the infrastructure to run Oracle E-Business Suite applications with Oracle Cloud Infrastructure Exadata in Google Cloud.

Artifact Registry - Gemini Cloud Assist for Artifact Registry is in Preview.

Bare Metal Solution - The command to connect to the interactive serial console of a server is changing on May 1, 2025.

BigQuery - Updated pricing, packaging, and setup guidance is now available for Gemini in BigQuery. You can now combine raster and vector data with the ST_REGIONSTATS geography function to perform geospatial analysis in BigQuery. You can now use the Apache Arrow format to stream data to BigQuery with the Storage Write API. Analytics Hub has been renamed BigQuery sharing. Dataplex Catalog has been renamed BigQuery universal catalog. BigQuery ML now offers a built-in TimesFM univariate time series forecasting model that implements Google Research's open source TimesFM model. You can now create, view, modify, and delete Apache Iceberg resources in BigQuery metastore. You can now connect BigQuery metastore to Apache Flink. BigQuery data preparation is generally available (GA). You can now create remote models in BigQuery ML based on Llama and Mistral AI models in Vertex AI. An updated version of JDBC driver for BigQuery is now available. Smart-tuning is now supported for materialized views when they are in the same project as one of their base tables, or when they are in the project running the query. BigQuery ML now uses dynamic token-based batching for embedding generation requests.

Bigtable - The Bigtable CQL client library for Java is available in Preview. The Cassandra-Bigtable proxy adapter, which lets you connect your Apache Cassandra-based applications to Bigtable, is available in Preview. The Bigtable Kafka sink, which lets you directly connect Apache Kafka and Google Cloud Managed Service for Apache Kafka, is now generally available (GA). Continuous materialized views for Bigtable are available in Preview. SQL support for Bigtable is generally available (GA), including an UNPACK feature that lets you read time series data in a tabular format. Logical views of Bigtable tables are available in Preview. The Bigtable Studio query editor is generally available (GA).

Chronicle - Premium parsers Specific high-volume parsers are now categorized as premium.

Chronicle Security Operations - Premium parsers Specific high-volume parsers are now categorized as premium.

Cloud Composer - The Airflow web server in Cloud Composer 3 requires at least 2 GB of memory when an environment is created or updated. All Cloud Composer environment's GKE clusters are set up with maintenance exclusions from March 27, 2025 to April 12, 2025.

Config Connector - Config Connector version 1.130.2 is now available. New Beta resources (direct reconciler) ApigeeEndpointAttachment ApigeeEnvgroupAttachment ApigeeInstanceAttachment ManagedKafkaTopic SecureSourceManagerInstance SecureSourceManagerRepository. New Fields GKEHubFeatureMembership Added spec.configmanagement.configSync.stopSyncing in version1.129. New Alpha resources (direct reconciler) ApphubApplication BackupDRManagementServer BackupDRBackupVault BackupDRBackupPlan BackupDRBackupPlanAssociation BatchJob BigLakeTable BigQueryReservation CodeDeployDeliveryPipeline DataplexLake DatastreamPrivateConnection DatastreamConnectionProfile DocumentAIProcessor GKEBackupBackupPlan GKEBackupRestorePlan NetAppBackupPolicy NotebooksEnvironment SpannerInstanceConfig VertexAIFeaturestore VMwareEnginePrivateCloud VMwareEngineNetwork VMwareEngineNetworkPeering VMwareEngineNetworkPolicy WorkflowExecution. Reconciliation Improvements Added support for direct reconciliation to more resources, with opt-in behaviour.

Contact Center AI Platform - Web SDK 2.24.4 patch is released This patch fixes a cross-site scripting vulnerability. Headless web SDK 3.6.3 is released Headless web SDK 3.6.3 fixes a cross-site scripting vulnerability.

Data Catalog - Dataplex Catalog has been renamed BigQuery universal catalog.

Database Migration Service - Gemini-powered auto-conversion is now available in Preview for all heterogeneous migration scenarios. Database Migration Service support for heterogeneous SQL Server to PostgreSQL migrations is now available in Preview.

Dataplex - Dataplex Catalog has been renamed BigQuery universal catalog.

Dataproc Serverless - Dataproc Serverless for Spark: Gemini Cloud Assist Investigations is available in Preview for the following runtimes: 1.1 1.2 2.2.

Dataproc - New Dataproc on Compute Engine subminor image versions: 2.2.52-debian12, 2.2.52-rocky9, 2.2.52-ubuntu22. Dataproc on Compute Engine: Fixed an issue with the retrieval of an Access token when using the ranger-gcs-plugin with 2.2 images.

Datastore - You can now use Query insights to view query performance metrics for your database.

Datastream - The Datastream API now supports streaming data to BigLake managed tables.

Document AI - Previous Custom Extractor versions pretrained-foundation-model-v1.0-2023-08-22 and pretrained-foundation-model-v1.1-2024-03-12 will be deprecated on April 9, 2025.

Anti Money Laundering AI - A bug was identified that can occasionally lead to parties appearing multiple times in prediction results.

Cloud Firestore - You can now use Query insights to view query performance metrics for your database. Firestore is now available on Database Center.

Gemini - Gemini Cloud Assist transforms cloud management and operations using the power of Gemini, placing AI at the center of your Google Cloud experience. Gemini Cloud Assist for Flow Analyzer is in Preview. Gemini Cloud Assist for Storage Insights is in Preview. Gemini Cloud Assist for Cloud Storage is in Preview. BigQuery data preparation is generally available (GA). Gemini Cloud Assist in App Hub is supported in Preview. Gemini in Firebase now provides AI assistance for the following Firebase products and features: Firebase Studio (Preview). Gemini Cloud Assist for Cloud Run is available in Preview. Gemini Code Assist tools are in Preview.

Identity-Aware Proxy - Preview: You can now enable IAP directly on your Cloud Run services without configuring load balancers.

Looker - Looker (Google Cloud core) and Looker (original) changes. Looker 25.6 is expected to include the following changes, features, and fixes: Expected Looker (original) deployment start: Monday, April 14, 2025 Expected Looker (original) final deployment and download available: Thursday, April 24, 2025 Expected Looker (Google Cloud core) deployment start: Monday, April 14, 2025 Expected Looker (Google Cloud core) final deployment: Monday, April 28, 2025. In the Chart Config Editor, you can save a configuration as a template so that you can reuse it in other visualizations or share it as a starting point for other users. The classification for the version, versions, and page_events API endpoints have been changed from "Admin" to "N/A" in System Activity queries. The Druid JDBC driver has been updated from 1.22.0 to 1.25.0. The Athena JDBC driver has been updated from 2.0.35.1000 to 2.1.5.1000. The Dremio JDBC driver has been updated from 4.5.0 to 25.2.0. The Spark Databricks JDBC driver has been updated from 2.6.34 to 2.7.1. The Exasol JDBC driver has been updated from 6.2.3 to 24.2.1. The Denodo JDBC driver has been updated from 8.8.0 to 9.1.3. The Trino JDBC driver has been updated from 402 to 468. An issue has been fixed where an Action Hub query could finish with a complete status even if the query failed. An issue has been fixed where sorting on a table visualization could fail to retrieve cached results, even if cached results were available for the query. An issue has been fixed where a dashboard tile could appear to load indefinitely if a user didn't have permission to the model. The file browser in the Looker IDE can now display files nested in 21 or fewer folders. An issue has been fixed where certain LookML validation errors could prevent Looker from successfully retrieving a list of models on the instance. If a user doesn't have an email address associated with their Looker account, the schedule dialog will not display the Send Test button. An issue has been fixed where an empty manifest file could cause the LookML Validator to display an error. An issue has been fixed where changing the subtotal column sort on dashboard tiles wouldn't properly update the sort order. An issue has been fixed where schedules to SFTP destinations could time out because of long SSH key generation times. An issue has been fixed where an embedded folder could still be loading content but not display a loading indicator. When uploading a JSON database authentication file to a connection, Looker now requires the file to be configured with the service_account type. An issue has been fixed where Looker would return a 500 error when it displayed a visualization with no results when the Grid Layout was set to By Row. Looker (original) only changes. A new Labs feature, Fast Dev Mode Transition, improves the performance of Development Mode on your instance by loading LookML projects in read-only mode until a developer clicks the Create Developer Copy button for the project. The New Database Connection Setup feature is now out of Labs and generally available. The Content Validator scoping feature is now generally available for customer-hosted Looker deployments (the feature is already available for Looker-hosted deployments). An issue has been fixed where embed users could save Looks to shared folders that they didn't have access to if the New Explore & Look Saving Labs feature was enabled. Looker (Google Cloud core) only changes. The New Database Connection Setup feature is now generally available. Looker (Google Cloud core) and Looker (original) changes. Looker has released version 1.4.2 of the Looker–Power BI Connector.

Cloud Monitoring - Application Monitoring lets you monitor the resources and infrastructure from the perspective of an App Hub application. Application Monitoring now supports app-enabled folders and App Hub host projects.

Cloud Interconnect - Cross-Site Interconnect is available in Preview.

Network Connectivity Center - IPv6 subnet exchange is generally available.

Network Intelligence Center - Gemini Cloud Assist for Flow Analyzer is in Preview.

Cloud Run - Gemini Cloud Assist in Cloud Run is supported in Preview. Direct VPC egress support for Cloud Run jobs is now generally available (GA). You can now configure Identity-Aware Proxy (IAP) for Cloud Run to secure your services with a single click from all ingress paths (in Preview). Configuring GPU in your Cloud Run service is now generally available (GA). Cloud Run Threat Detection is available in Preview.

Security Command Center - Model Armor and GKE integration Model Armor now enforces security policies uniformly on generative AI inference traffic using a traffic extension. Cloud Run Threat Detection is available in Preview.

Sensitive Data Protection - If you set InfoType.version to latest when including the MAC_ADDRESS infoType in your InspectConfig, Sensitive Data Protection will now include MAC_ADDRESS_LOCAL findings as type MAC_ADDRESS in the scan results.

Service Extensions - The Google Kubernetes Engine (GKE) Gateway supports using extensions to add custom logic into the load balancing processing path. You can configure Model Armor with Service Extensions to protect AI workloads on supported Application Load Balancers.

Service Health - You can ask Gemini Cloud Assist about active incidents, their impact, and updates You can view incidents related to your App Hub applications in Cloud Hub.

Cloud Spanner - You can use Gemini assistance to help you use system insights to optimize and troubleshoot Spanner resources. Spanner offers Cassandra compatibility with API support and new migration tools allowing seamless lift-and-shift migrations of Cassandra applications.

Cloud SQL MySQL - Cloud SQL Enterprise Plus edition now supports a new machine series called the C4A machine series, which provides optimized price-performance and delivers predictable high performance for high demand Cloud SQL workloads. Query insights for Cloud SQL Enterprise Plus edition is now generally available (GA) for your Cloud SQL Enterprise Plus edition for MySQL instances. Cloud SQL for Enterprise Plus edition supports AI-assisted troubleshooting.

Cloud SQL Postgres - Cloud SQL Enterprise Plus edition now supports a new machine series called the C4A machine series, which provides optimized price-performance and delivers predictable high performance for high demand Cloud SQL workloads. Query insights for Cloud SQL Enterprise Plus edition is now generally available (GA) for your Cloud SQL Enterprise Plus edition for PostgreSQL instances. Cloud SQL for Enterprise Plus edition supports AI-assisted troubleshooting.

Cloud SQL SQL Server - Query insights for Cloud SQL Enterprise edition and Cloud SQL Enterprise Plus edition is now generally available (GA) for Cloud SQL for SQL Server. Cloud SQL now supports the Enterprise Plus recommender.

VPC Service Controls - General availability support for the following integration: Database Center.

Virtual Private Cloud - When you create a Private Service Connect endpoint to connect to a regional endpoint of a supported service, you can use the public hostname in your configuration—for example, spanner.me-central2.rep.googleapis.com.