All news in category “Vendor and Hyperscaler Watch”

🛠️ Microsoft has fixed a known issue that caused systems running Windows Server 2019 and 2022 to unexpectedly upgrade to Windows Server 2025. The problem was first acknowledged in September 2024 after widespread reports from administrators, and Microsoft says it has re-enabled the in-place upgrade offer via the Settings app. Microsoft previously cited third-party update management configuration, while some vendors said the root cause was a procedural error on Microsoft's side.

🔎 Mallory has launched an AI-native threat intelligence platform that converts global threat telemetry into prioritized, evidence-based cases tailored to an organization’s environment. The SaaS offering monitors thousands of sources, contextualizes findings against actual attack surfaces, and integrates with existing tools to automate hunt, detection, and exposure management workflows. It emphasizes actionable answers over alerts and supports Claude Code, MCP, APIs, and a modern UI for extensibility.

🚀 Amazon has added EC2 P6-B300 instances to the AWS GovCloud (US‑East) Region. The p6-b300.48xlarge configuration provides 8x NVIDIA Blackwell Ultra GPUs with 2.1 TB of high-bandwidth GPU memory, 6.4 Tbps EFA networking, 300 Gbps ENA throughput, and 4 TB of system memory. P6-B300 delivers ~2x networking, 1.5x GPU memory and 1.5x FP4 TFLOPS vs P6-B200, targeting training and deployment of large trillion-parameter foundation models and LLMs.

🚀 AWS Transform is now accessible from within Kiro and Visual Studio Code via an AWS Transform Power and an extension, enabling developers to start, run, and monitor agentic migration and modernization jobs directly from their IDEs. The platform offers custom transformations and AWS-managed playbooks for common upgrades — language version bumps and SDK migrations — and supports running agents across thousands of repositories. Job state and context are synchronized across the web console, CLI, and IDEs, and AWS Transform custom is currently available in US East (N. Virginia) and Europe (Frankfurt).

🔐 AWS Secrets Manager now supports hybrid post-quantum TLS key exchange using ML-KEM (a module-lattice-based KEM) to secure secret retrieval. The capability is automatically enabled in Secrets Manager Agent (v2.0.0+), Lambda Extension (v19+), and Secrets Manager CSI Driver (v2.0.0+); supported SDKs include Rust, Go, Node.js, Kotlin, Python (OpenSSL 3.5+), and Java v2 (v2.35.11+). No code or configuration changes are required for up-to-date clients except Java v2. You can verify hybrid key exchange in CloudTrail GetSecretValue events by checking the tlsDetails field for the X25519MLKEM768 algorithm.

🚚 AWS now lets customers deliver AWS Data Exports — including Cost and Usage Report 2.0 (CUR 2.0), FOCUS, Cost Optimization Recommendations, and Carbon Emissions reports — directly into an authorized Amazon S3 bucket in another AWS account. This removes the need to replicate billing data across accounts and eliminates duplicate storage costs by allowing exports to target a destination bucket when created. The destination account owner controls which source accounts may deliver data via S3 bucket policies. The capability is available in all commercial AWS Regions except AWS GovCloud (US) and China Regions.

🔍 BigQuery Graph and Kineviz GraphXR deliver an integrated workflow to extract, model, and visualize knowledge from unstructured enterprise data without separate graph databases or heavy ETL. By performing text extraction, Gemini-powered inference, and graph creation inside BigQuery, organizations preserve provenance and avoid data duplication. GraphXR connects live to the graph for low-code, interactive analysis, evidence-linked tracing, and dashboards that update as graph views evolve.

🔍 BigQuery Graph is now available in preview, offering an integrated, serverless graph analytics capability within BigQuery that scales to billions of nodes and edges. It provides an intuitive graph query experience using a GQL dialect aligned with the ISO GQL standard and full interoperability with SQL, removing the need to copy or move data. The preview includes vector and full-text search, notebook visualizations, and federation with Spanner Graph for combined real-time and historical analysis, aimed at use cases such as fraud detection, drug discovery, and supply-chain analysis.

🔒 Google outlines an AI-focused security strategy for public sector organizations, emphasizing agentic SOCs powered by Gemini agents and Mandiant frontline expertise. The post summarizes 2026 threat trends — compressed attack cycles, prolonged nation-state access, rising voice phishing, and emerging shadow agents — and stresses integrated visibility across code, cloud, and runtime via Security Command Center. It highlights operational gains such as Connecticut reducing investigations from months to hours and previews demonstrations at Google Cloud Next.

🔗 Google Cloud introduces a unified graph solution that pairs Spanner Graph for operational (OLTP) workloads with BigQuery Graph for analytical (OLAP) queries, enabling developers and analysts to work against a consistent GQL schema without duplicating data. Both platforms support integrated table-to-graph mapping, mixed GQL/SQL queries, built-in vector and full-text search, and AI integrations to power real-time applications and large-scale historical analysis. The solution also offers cross-system workflows via Data Boost (query Spanner from BigQuery), reverse ETL exports to Spanner, and visualization integrations with partners like Kineviz and Linkurious to accelerate investigations and insights.

🔐 Microsoft has introduced a temporary fast-track to reinstate accounts suspended from the Windows Hardware Program after developers reported being locked out without prior notice. The process asks affected partners to open a support case, provide a clear business justification, and resolve outstanding compliance requirements before full access is restored. Microsoft also provided guidance on correct sign-in and alternative support contacts to address workflow issues.

☁️ Azure announces the general availability of smart tier for Blob and Data Lake Storage, a fully managed automated tiering service that continuously optimizes object placement across hot, cool, and cold tiers. It evaluates last-access timestamps—objects idle 30 days move to cool and after 60 more days move to cold—and promotes data back to hot on access. Enable during account creation or switch existing zonal accounts to start optimizing automatically.

🚀 AWS has increased Amazon EBS performance for EC2 C8gn, M8gn, and R8gn instances in 48xlarge and metal-48xl sizes. With enhancements to the AWS Nitro System, EBS bandwidth doubles from 60 Gbps to 120 Gbps and IOPS doubles from 240,000 to 480,000. New launches receive the upgrade at no extra cost; running instances can enable it by stopping and starting. The change is available in all regions where these instance types are generally available.

🛡️ Google has integrated a Rust-based DNS parser into the modem firmware for Pixel 10, marking the first Pixel modem component written in a memory-safe language. The change aims to eliminate a broad class of memory-safety bugs in DNS handling, using the hickory-proto crate adapted for embedded use and a custom cargo-gnaw tool to manage dependencies. The Rust implementation exposes a C API and dispatches existing C functions to update in‑memory structures.

🔐 Cloudflare announces updates to secure non-human identities—agents, scripts, and third-party tools—by enhancing credential detection, OAuth visibility, and resource-scoped RBAC. New scannable token formats (with identifiable prefixes and checksums) and integration with GitHub Secret Scanning enable rapid verification and automated revocation of leaked tokens. Cloudflare One DLP extends prevention across network, email, SaaS, and AI traffic. The Dashboard now surfaces connected OAuth apps and permissions to simplify review and revocation.

🔐 Cloudflare is launching Managed OAuth for Cloudflare Access in open beta, enabling agents that speak OAuth 2.0 to authenticate to internal apps with a single click. When enabled, Access acts as the authorization server and uses the www-authenticate header to point agents to the /.well-known/oauth-authorization-server. Agents can dynamically register (RFC 7591), perform PKCE (RFC 7636), and receive JWTs to act on behalf of users, removing the need for static service accounts.

🔒 Cloudflare Mesh provides a developer-friendly private network that unifies access for users, devices, and AI agents across clouds and the Cloudflare edge. Integrated with Cloudflare One, Mesh uses the Cloudflare One Client and Mesh nodes to route bidirectional, many-to-many traffic with built-in Gateway policies, DNS filtering, device posture checks, and DLP. It supports Workers VPC bindings and the Agents SDK so serverless agents and Durable Objects can securely reach private services, with a free tier for up to 50 nodes and 50 users.

🚀 Amazon SageMaker JumpStart now includes NVIDIA’s Nemotron-3-Super-120B and the Qwen3.5 family (9B and 27B), giving customers turnkey access to foundation models optimized for agentic reasoning, multilingual coding, and advanced instruction following. Nemotron-3-Super-120B employs a hybrid LatentMixture-of-Experts architecture with Mamba-2 and MoE layers to support collaborative agents and high-volume automation such as IT ticket triage and cybersecurity workflows. The Qwen3.5-9B prioritizes efficiency for resource-constrained environments, while Qwen3.5-27B offers deeper contextual and multimodal reasoning for large-scale document processing and complex scenarios. Users can deploy these models directly from the JumpStart catalog or programmatically via the SageMaker Python SDK.

🔧 Amazon OpenSearch Serverless now supports Derived Source, a feature that reconstructs the _source field on demand using indexed values instead of storing a separate copy. This reduces storage consumption for collections such as time-series and log analytics that contain many indexed fields. You can enable Derived Source at the index level when creating or updating mappings. The feature is available today in all AWS Regions where OpenSearch Serverless is supported.

🚀 Amazon Redshift now accelerates top-k queries (those using ORDER BY with LIMIT) by skipping irrelevant data blocks and keeping only the K best candidate rows in memory. The engine reorders and selectively reads blocks based on the ORDER BY column's min/max values so that when data is fully or partially sorted it reads only the minimal blocks needed rather than scanning entire tables. This optimization is available at no extra cost in patch release P199 and applies automatically to eligible queries without any rewrites or configuration changes.