All news with #google cloud tag

🚀 Cloud Run now supports NVIDIA RTX PRO 6000 Blackwell GPUs in preview, enabling serverless deployment of large inference models such as Gemma 3 27B and Llama 3.1 70B. The GPUs provide 96GB vGPU memory, 1.6 TB/s bandwidth and support for FP4 and FP6 precision. Cloud Run pre-installs drivers, offers rapid GPU startup and autoscaling to zero, and integrates with Cloud Storage and IAP for production use.

🔍 Conversational Analytics in BigQuery (preview) brings an AI-powered reasoning agent into BigQuery Studio, enabling users to query, visualize, and forecast directly with natural language. The agent generates and executes SQL grounded in your schema, metadata, and verified queries, and it exposes the SQL and reasoning behind each answer to build trust. Security, governance, and audit logging are enforced by BigQuery’s compliance controls, and the feature also supports unstructured data and API integration for custom agents.

🚀 In 2025, Spanner became a first-class database for AI and multi-model workloads, integrating vectors, graphs, and text search to provide richer context for generative applications. Google expanded AI integrations including ML.PREDICT, Vertex RAG Engine support, an Agent Development Kit, and ANN search with ScaNN to speed semantic retrieval. The release also bridged operational and analytical silos with a columnar engine and tighter BigQuery/Iceberg integrations, simplified Cassandra and MySQL migrations, boosted price-performance, and strengthened enterprise safeguards such as drop protection and default backup schedules.

🔌 AlloyDB now offers managed connection pooling for PostgreSQL as a generally available capability, reducing connection overhead and improving scalability. The service-managed pooler listens on port 6432 and reuses backend connections to cut latency and resource churn, while Google Cloud handles setup, configuration, and maintenance. Choose Transaction or Session pooling based on compatibility needs and configure sizes via Console, gcloud, or API. Customers report support for 3x more clients and up to 5x higher transactional throughput.

🚀 Fastweb + Vodafone rebuilt its Customer 360 platform on Google Cloud to deliver unified, real-time customer insights using Spanner, BigQuery, and Gemini. The team replaced brittle, high-maintenance integrations with a connected platform that enables low-latency serving, centralized governance, and AI-assisted engineering workflows. Spanner’s native BigQuery integration sped migrations and simplified monitoring, while Apigee exposes consistent data to call centers, digital channels, and partners.

🚀 This post describes two validated deployment recipes for serving large Mixture-of-Experts (MoE) models on Google Cloud's A4X machines using NVIDIA Dynamo. The recipes provide throughput- and latency-optimized configurations that exploit the 72‑GPU GB200 NVL72 rack fabric, WideEP/DeepEP parallelism, global KV cache, and GKE-aware rack-level scheduling. Performance validation reports >6K tokens/sec/GPU for the throughput recipe and a 10ms median inter-token latency for the latency-optimized recipe.

🚀 Google Cloud has refreshed the Rapid Migration and Modernization Program (RaMP) to incentivize cloud migrations with service credits tied to incremental usage and funded partner and professional services. The program offers enhanced rewards for advanced workloads—SAP, Oracle, VMware, NetApp and data analytics—to help offset higher technical costs. RaMP is positioned to reduce technical debt, accelerate AI readiness by making data accessible to Vertex AI and Gemini, and provide a funded path for assessment and implementation.

⚠️ A recent Pentera investigation discovered nearly 2,000 intentionally vulnerable security-testing web applications (DVWA, OWASP Juice Shop, Hackazon, bWAPP) exposed on the public internet, often running from overly privileged cloud accounts on AWS, GCP and Azure. Attackers exploited these instances to deploy crypto miners, install webshells and create persistence mechanisms, then pivot to sensitive cloud resources. Affected vendors including Cloudflare, F5 and Palo Alto Networks were notified and remediated issues. Pentera recommends inventories, isolation of test systems, enforcement of least-privilege IAM, and elimination of default credentials.

🚀 Google Cloud has launched a new Bangkok (asia-southeast3) region to deliver low-latency, high-performance cloud services while enabling local data residency under Thailand’s PDPA. The region is part of a USD $1 billion investment and is expected to generate THB 1.4 trillion (US$41 billion) in economic value over five years and support roughly 130,000 jobs annually. It offers certified security controls (ISO/IEC, PCI DSS, SOC), default encryption, customer-managed keys, and direct access to Vertex AI, enterprise Gemini, and generative models to accelerate local AI adoption.

🚀 This post introduces Gemini 3 Flash, Google’s low-latency, cost-efficient model in the Gemini 3 family, optimized for advanced reasoning, multimodal understanding, and agentic workflows. It guides developers through obtaining an API key from Google AI Studio and configuring it for local use or environment-based invocation. The article demonstrates interactive prompt testing in the Playground, explains toggles like Structured outputs and Thinking level, and shows how to export language-specific sample code via the "Get code" feature to run with the Google GenAI SDK.

🚀Palo Alto Networks partnered with Google Cloud to replace a brittle single-tenant data pipeline model with a unified, multi-tenant Unified Data Platform powered by Dataflow, Pub/Sub and BigQuery. The migration consolidated more than 30,000 pipelines into a shared, autoscaling platform that processes billions of events daily. The change delivered roughly 30% compute cost savings, faster onboarding, and reduced operational overhead, enabling engineers to refocus on analytics and threat detection.

🔐 Tom Curry and Anton Chuvakin from Google Cloud’s Office of the CISO present practical guidance for implementing the Secure AI Framework (SAIF) on Google Cloud. The piece emphasizes three operational principles: treat data as the perimeter, treat prompts like code, and require identity propagation for agentic AI. It maps 15 common AI risks to controls and highlights concrete tools and patterns—IAM, Dataplex, Vertex AI, Model Armor, Gemini, Apigee, and the Agent Development Kit—to operationalize SAIF.

🤖 This recap of the Agent Factory holiday special summarizes practical guidance on model fine-tuning, with a focus on reinforcement learning (RL) and Google’s TPU infrastructure. Hosts Shir Meir Lador and Don McCasland speak with Kyle Meggs from the TPU Training Team about when to fine-tune, the distinction between pre‑training, SFT, and RL, and why specialized workloads benefit from hosted solutions like MaxText on TPUs. The post also demonstrates a GRPO demo using Pathways, vLLM, and Tunix components to show RL at scale.

🔗 Google Cloud describes work to enable gRPC as a native transport for the Model Context Protocol (MCP), offering an alternative to JSON-RPC transcoding for organizations that already use gRPC. Native gRPC eliminates the need for transcoding gateways and preserves existing tooling, while delivering lower latency, smaller Protobuf-encoded payloads, and full-duplex streaming. The MCP core maintainers agreed to pluggable transports in the SDK, and Google Cloud will contribute a community-backed gRPC transport package to promote consistent, interoperable deployments.

🔍 Hackensack Meridian Health used VPC Flow Logs and Flow Analyzer to obtain precise, end-to-end visibility of Cloud Interconnect traffic before a major Google Cloud network migration. They enabled VLAN-attachment flow logs, aggregated ingress/egress flows (IPs, ports, bytes, timestamps), and organized results into sankey diagrams mapping data center → region → VPC → application. This process revealed critical flows early and shortened incident detection to 3 minutes and resolution to 5 minutes, materially de-risking the cutover.

🔍 FINRA partnered with Google Cloud to adopt the DORA metrics and a data-first DevOps approach to shorten lead times and modernize its software lifecycle. A DORA workshop revealed lengthy User Acceptance Testing (UAT) cycles as a primary bottleneck, enabling a multi-million-dollar business case for a dedicated sandbox to accelerate testing and deployment. The initiative standardized DORA across teams and targets full adoption within the year.

🔍 Google Cloud has enhanced Gemini CLI telemetry with pre-configured Google Cloud Monitoring dashboards that provide immediate visibility into adoption, usage patterns, and performance. By exporting data via OpenTelemetry, teams can use out-of-the-box visualizations or analyze raw logs and metrics to build custom views. Setup is simplified through direct GCP exporters and a three-step flow—project ID, authentication and IAM roles, and updating .gemini/settings.json—so telemetry can be live quickly.

🔒 Check Point now supports Google Cloud Network Security Integration, offering a nondisruptive approach to deploying cloud firewalls that minimizes downtime and avoids performance degradation. The integration enables organizations—particularly in regulated sectors such as financial services, healthcare, and government—to scale hybrid network security while preserving latency and throughput. It simplifies deployment, centralizes policy management, and helps maintain compliance without rearchitecting existing networks.

⚡ Cloud SQL for MySQL Enterprise Plus now includes optimized writes, an automated runtime tuning suite that adjusts MySQL configuration and I/O behavior to reduce write latency and increase throughput. Enabled by default on Enterprise Plus instances, the feature implements adaptive purge, adaptive I/O limits, sharded I/O, faster REDO recovery, and adaptive buffer-pool warmup. Google provides a reproducible sysbench benchmark and reports up to 3x write throughput improvements versus the Enterprise edition, with results varying by machine type and workload.

🚗 Google Cloud has joined the Automotive Information Sharing and Analysis Center as an Innovator Partner, pledging experts and resources to bolster vehicle and supply-chain cybersecurity. The partnership will bring threat intelligence and incident response expertise — including insights from Mandiant — to help members anticipate, mitigate, and respond to attacks against cloud-connected, software‑defined vehicles and Industry 4.0 environments. Google cites a $10 billion cybersecurity investment over five years as part of its broader commitment.