All news in category “Vendor and Hyperscaler Watch”

🚀 Starting today, Aurora PostgreSQL can directly query operational data alongside data stored in data lakes in Apache Iceberg and Parquet formats using your existing PostgreSQL applications and tools. You can create PostgreSQL foreign tables that reference data in Amazon S3, Amazon S3 Tables, or AWS Glue Data Catalog, and queries against those foreign tables use DuckDB’s high-performance engine embedded in PostgreSQL. The capability supports external Iceberg REST Catalog–compatible catalogs federated via AWS Glue Data Catalog, and data can be materialized into native Aurora tables for low-latency workloads. It is generally available on Aurora PostgreSQL versions 17.11, 18.6 and higher across AWS commercial and GovCloud (US) Regions at no extra charge.

🔍 AWS CloudTrail has launched Event Coverage, a console feature that displays data plane event logging coverage across accounts and organizations. It highlights which services and resource types have data event logging enabled and which do not, helping teams identify logging gaps without manual scans. The dashboard consolidates supported data event sources and lets users subscribe to data events directly to close coverage gaps quickly.

🔍 Students in technical fields often worry their education lacks practical skills and fear being unprepared for the job market. Experts say this anxiety is normal and can be eased by building identity capital — practical experience, skills, and connections — through internships, courses, and adjacent learning. With digital threats rising and human error driving incidents, basic cybersecurity knowledge is increasingly essential for employers and graduates alike.

📊 Amazon Quick adds a new Hierarchy Filter control that consolidates multiple stacked filters into a single, guided dropdown for drilling through related dimensions like Region, Country, and City. The filter supports up to five levels arranged from broadest to most detailed, allowing readers to expand each level, narrow selections progressively, and auto-select parent relationships. Authors enable it by adding a filter, choosing Hierarchy filter type, and ordering fields parent-to-child. The feature is available today in all AWS Regions supporting Amazon Quick.

🟢 Amazon Quick now enables applications to consume live data directly from Quick datasets, keeping KPIs, charts, tables, and insights continuously up to date. These apps support interactive business workflows—handling requests, approvals, and triggering next steps—while respecting each user’s existing Amazon Quick row- and column-level permissions. Builders create apps by naming a dataset and describing requirements in plain language to the chat agent, simplifying integration of current pipeline numbers into portals and workflows. The feature is available in all supported regions of Amazon Quick apps.

🔧 Microsoft Azure outlines how it manages the full lifecycle of hyperscale hardware to balance increased compute density with responsible reuse and recycling. The company describes design, operation, secure decommissioning, and repurposing through its global Circular Centers, now expanded to North America, Europe, and Asia Pacific. Azure reports a 92% reuse and recycling rate and highlights automation and AI-driven disassembly to scale circular operations and support Zero Waste goals.

🔒 Microsoft will begin enforcing stricter Content Security Policy (CSP) protections for Entra ID sign-ins starting mid-October 2026, allowing only scripts from trusted Microsoft CDN domains. The rollout will complete by late October 2026 and aims to block external script injection and cross-site scripting risks during browser-based authentication. Enterprise customers are urged to remove or test browser extensions and code-injection tools to avoid sign-in disruptions. MSAL and API-based flows are not affected because CSP applies only to browser-based sign-ins.

🌐 Cloudflare’s Impact programs have reached $100 million in donated services, protecting thousands of organizations including newsrooms, civil society groups, governments, and schools from cyberattacks. Originally launched as free services, programs like Project Galileo now defend over 3,500 domains in 120+ countries and blocked billions of attacks in 2025. Cloudflare’s efforts have evolved to include developer support, hackathons for journalists, and tools to help public interest groups adapt to AI-driven threats.

🔍 Cloudflare has redesigned its domain search and Registrar experience to be faster, clearer, and less cluttered. The new interface shows the exact query across 420+ extensions, updates results as you type, and supports sorting, filtering, and transparent pricing including renewal costs. Developers and agents can use the expanded Registrar API, sandbox, and cf CLI to search, register, or transfer domains programmatically. The system balances speed and accuracy via prepared datasets, DNS checks, and live registry lookups while scaling on the Cloudflare developer platform.

🔍 Cloudflare’s User Insights now adds context to AI usage by grouping conversations by task and highlighting when a model is more capable than required. The update surfaces which users, agents, and applications drive overuse and links task categories, token and latency metrics, and conversation turns to cost. These signals power a new Potential Savings view and the Auto Router beta to help teams route requests to appropriate, cost-effective models.

📰 Cloudflare has introduced Pay Per Use, a beta program that lets AI companies offer to pay publishers for specific downstream uses of their content. Publishers can review offers, accept or decline, and monitor usage and earnings in the Cloudflare dashboard. Buyers report uses through a single API, Cloudflare aggregates reports, bills buyers, and pays publishers monthly. The model complements Pay Per Crawl and aims to convert downstream AI uses into measurable revenue.

🔒 Today Cloudflare announced the Monetization Gateway is available in a closed beta, enabling sellers to charge agents per request for web resources, APIs, tools, and datasets. The gateway uses HTTP 402 to inline payment negotiations with resource requests and settles transactions using USDC on the Base blockchain via Coinbase's x402 Facilitator. Several customers — including AI Gateway, Ceramic.ai, Stocktwits, and API2PDF — are already using consumption-based pricing and variable billing in production. U.S.-based sellers and buyers can request access through the Cloudflare Dashboard.

🔥 Cloudflare announces Issues, a built-in error monitoring feature for Cloudflare Workers now in open beta. With one-line configuration, Issues captures uncaught exceptions, failed invocations, 5xx responses, console output, and stack traces without additional SDKs. It groups related failures, shows trends and context, and can send structured diagnostics to agents via Automations for investigation and remediation. Use the Workers OpenTelemetry API to attach application identifiers and the cf CLI to inspect issues and create automations.

🧭 Cloudflare announces Auto Router in public beta via AI Gateway, automatically routing requests to the most cost-effective capable model without requiring user selection. Early internal use shows up to 30% cost savings versus always using frontier models like OpenAI Sol and Anthropic Opus. Auto Router classifies requests by task and dimensions, scores candidate models by quality and token costs, and accounts for cache-read/write penalties when switching models. The router returns a ranked list and falls back if a provider cannot serve the request.

🛠️ Today Google Cloud announces the general availability of Data Agent Kit, a free set of Model Context Protocol (MCP) tools and Google-authored agent skills that let coding agents access and act on Google Cloud data products. The GA release adds support for BigQuery Graph, Bigtable, and Managed Service for Apache Spark in Lakehouse workflows, plus numerous quality-of-life and performance improvements. The kit integrates with IDEs, Antigravity, Claude Code, Codex, Cloud Shell, and Cloud Workstations, enabling agents to inspect schemas, run queries, read job logs, and orchestrate pipelines using customers' IAM permissions.

🔎 Cloudflare describes a rapid shift: more than half of web traffic is now generated by AI agents rather than humans. The company explains how agents differ from traditional crawlers and why blanket blocking is insufficient. Cloudflare outlines new controls, cryptographic bot identity, and products like Pay Per Use and Monetization Gateway to let publishers charge verified buyers. These tools aim to restore revenue, provide transparency, and create a payable, discoverable Agentic Internet.

🛠️ Cloudflare rearchitected Containers to better serve on-demand agent sandboxes by allowing code to choose sandbox images and instance types at runtime. The new durable_object scheduling policy and redesigned runtime yield a 6x faster median startup and improved burst performance, while filesystem snapshots enter public beta. Durable Objects now control containers more directly via the native ctx.container API, simplifying rollout and enabling richer, per-task environment decisions.

🔔 Partner Revenue Measurement (PRM) now supports multi‑Partner Resource Tagging so multiple AWS Partners can receive revenue attribution for the same AWS resource. Previously, a single aws-apn-id tag limited attribution to one Partner. The new tag format, aws-apn-id-, lets each contributing Partner tag resources they work on, enabling shared attribution while keeping the legacy tag functional. This capability is available today in all commercial AWS regions and appears under the “Resource Tagging” method in the Attributed Revenue Dashboard.

🔒 This post compares two reference networking architectures for AI inference model serving: one tailored for Google Kubernetes Engine (GKE) and one for mixed or alternative backends. It explains a common control-plane pattern using Private Service Connect, optional Apigee, and Model Armor as a centralized entry point for secure, private inference calls. The GKE design adds a specialized GKE Inference Gateway, inference pools, and replica sets for GPU/TPU workloads. The multi-backend design uses a regional internal Application Load Balancer, a Cloud Run payload processor service extension to inject model headers, and Network Endpoint Groups to route to heterogeneous backends.

🔒 AWS IAM Identity Center's Identity Store APIs now accept ARNs for users, groups, group memberships, and identity stores wherever resource IDs were previously required. This additive change means existing integrations keep working while allowing callers to supply either resource IDs or full ARNs directly. Responses still return resource IDs and malformed or wrong-type ARNs result in a ValidationException. The feature is available in all Regions where IAM Identity Center is offered at no extra cost.