All news in category “Vendor and Hyperscaler Watch”

🚀Amazon Web Services has launched EC2 C7i-flex instances in the Middle East (UAE), offering up to 19% better price performance versus C6i. Powered by AWS-exclusive 4th generation Intel Xeon Scalable (Sapphire Rapids) custom processors and priced about 5% below C7i, these instances cover common sizes from large through 16xlarge. They target web and application servers, caches, databases, Apache Kafka, Elasticsearch and other compute-intensive workloads that don’t fully utilize all vCPUs. For sustained heavy-CPU needs or very large configurations, customers can continue to use standard C7i instances.

🚀 Amazon Web Services has added High Memory U7i instances to AWS GovCloud, offering 12TiB (u7i-12tb.224xlarge) and 16TiB (u7in-16tb.224xlarge) in GovCloud (US-West) and 24TiB (u7in-24tb.224xlarge) in GovCloud (US-East). These 7th‑generation instances use custom 4th‑generation Intel Xeon Scalable (Sapphire Rapids) processors, provide 896 vCPUs and DDR5 memory, and support ENA Express. The u7i-12tb delivers up to 100Gbps network and EBS throughput while the 16tb and 24tb variants deliver up to 200Gbps, making them well suited for mission‑critical in‑memory databases like SAP HANA, Oracle, and SQL Server.

🌐 Amazon Web Services now supports IPv6 addresses for AWS PrivateLink Gateway and Interface VPC endpoints for Amazon S3. To enable IPv6 connectivity on new or existing S3 endpoints, set the IP address type to IPv6 or Dualstack; S3 will update route tables for gateway endpoints and provision ENIs with IPv6 for interface endpoints. IPv6 for S3 VPC endpoints is available in all AWS Commercial Regions and AWS GovCloud (US) Regions at no additional cost, and can be configured via the Console, CLI, SDK, or CloudFormation.

📈 Amazon Web Services announced that the Amazon CloudWatch Agent can now collect shared memory utilization metrics from Linux hosts running on Amazon EC2 or in on‑premises environments. This complements existing memory metrics (free, used, cached) and captures memory used by large enterprise databases and in‑memory applications. Administrators can enable the feature in the agent configuration file to obtain accurate total memory usage for sizing and optimization. The capability is available in all commercial and AWS GovCloud (US) Regions; CloudWatch custom metrics pricing applies.

🔔 Amazon SageMaker Unified Studio now delivers real-time notifications for data catalog activities, including new dataset publications, metadata changes, subscription requests, comments, and access approvals. Alerts are surfaced via a bell icon on the project home page and through a notification center that shows a recent list and a full, filterable tabular view by catalog, project, and event type. The feature is available in all regions where SageMaker Unified Studio is supported.

🔐 AWS Private CA now supports the post-quantum digital signature algorithm ML-DSA (NIST FIPS 204), enabling organizations to create CAs and issue certificates designed to resist quantum attacks. The feature lets you test certificate issuance, identity verification, and code signing using ML-DSA, and supports CRLs and OCSP responders. Availability spans all commercial AWS Regions, AWS GovCloud (US), and China Regions to help teams begin transitioning PKI toward post-quantum cryptography.

🔒 CrowdStrike has been named the Overall Leader in the 2025 KuppingerCole Leadership Compass for Identity Threat Detection and Response, achieving top placement across Product, Innovation, Market, and Overall Ranking. The report cites Falcon Next-Gen Identity Security for its cloud-native design, AI/ML-driven detections, behavioral analytics, and automated identity-centric response. KuppingerCole highlights unified visibility across Active Directory, Entra ID, Okta, Ping, AWS IAM and SaaS via Falcon Shield, and notes deep integrations with XDR, SIEM, SOAR, IdP, IGA, PAM, and ITSM to accelerate detection and remediation for human, non-human, and AI agent identities.

🔔 The Windows 11 Start menu has received its first major redesign since 2021 and is rolling out with the November 11 Patch Tuesday update. The new Start is scrollable and places the All apps list on the main screen, offering a categorized view (groups built locally from a JSON file) and a classic A‑to‑Z grid. The UI adapts column counts to screen size, lets you hide the Recommended feed via Settings > Personalization > Start, and is included in Build 26200.7019 and 26100.7019 or newer though it may not enable immediately after updating.

🔁NAKIVO has released Backup & Replication v11.1, expanding disaster recovery and MSP capabilities and adding five interface languages—French, Italian, German, Polish and Chinese. The update brings major Proxmox VE improvements, including Flash VM Boot, VM replication and template backup/recovery, automated backup verification with screenshots, direct tape recovery, and Exchange/SQL log truncation. It also introduces MSP Direct Connect to remove client-side port changes, Real-Time Replication for VMware with automated IO Filter and Journal Service installation, and granular folder- and volume-level backups for Windows and Linux physical machines with encryption, immutability and air-gapping options.

🔁 Microsoft is testing a faster version of Quick Machine Recovery (QMR) in Windows 11 that runs a one‑time scan in the Windows Recovery Environment to more quickly identify and apply fixes for systems that fail to boot. When WinRE launches QMR it connects to the internet to upload crash data so administrators can remove problematic updates or adjust settings remotely. The update also lets administrators and users toggle Smart App Control from Windows Security without performing a clean OS install, and is currently available to Insiders on Dev and Beta via Build 26220.7070 (KB5070300).

🌐 Amazon VPC Lattice now lets resource owners assign a custom domain name to a resource configuration, enabling layer‑4 access to databases, clusters and TLS‑based endpoints across VPCs and accounts. Owners specify a custom domain and share the resource configuration; VPC Lattice then provisions and manages a private hosted zone in the consumer VPC so consumers can resolve and access the resource using that domain. Resource owners may use AWS, customer‑owned, or third‑party domains, and consumers can exercise granular controls over which domains VPC Lattice manages. The feature is available at no additional cost in Regions where VPC Lattice resource configuration is offered.

🔒 Ericsson has implemented a global data governance framework using Dataplex Universal Catalog on Google Cloud to ensure data integrity, discoverability, and compliance across its Managed Services operation. The program standardized a business glossary, automated quality checks with incident-driven alerts, and visualized column-level lineage to support analytics, AI, and automation at scale. It balances defensive compliance with offensive innovation and embeds stewardship through Ericsson’s Data Operating Model.

🧭 This post explains when to use sub-agents versus packaging agents as tools when building multi-agent systems with Google's Agent Development Kit (ADK). It contrasts agents-as-tools — encapsulated, stateless specialists invoked like deterministic function calls — with sub-agents, which are stateful, context-aware delegates that manage multi-step workflows. The guidance highlights trade-offs across task complexity, context sharing, reusability, and autonomy, and illustrates the patterns with data-agent and travel-planner examples to help architects choose efficient, scalable designs.

🔍 AlloyDB AI launches two preview features—Auto Vector Embeddings and Auto Vector Index—that let teams convert operational databases into AI-native stores using simple SQL. Auto Vector Embeddings generates and incrementally refreshes vectors in-database, batching calls to Vertex AI and running as a background process. The Auto Vector Index (ScaNN) self-configures, self-tunes, and maintains vector indexes to accelerate filtered semantic search and reduce ETL and tuning overhead for production workloads.

🚀 Deploy the official n8n Docker image to Cloud Run in minutes to run scalable, serverless AI workflows. Cloud Run scales from zero and persists data in Cloud SQL while you only pay for active usage. The post shows how to call Gemini as the agent LLM and optionally connect workflows to Google Workspace via OAuth for Gmail, Calendar, and Drive. For production, follow the n8n docs to add Secrets Manager, Cloud SQL, and Terraform-based deployment.

🇪🇺 Google Cloud has formed a new European Advisory Board to provide strategic counsel on regulatory, product, and market priorities and to help customers navigate complex European requirements. The board unites leaders from technology, finance, retail, and public service, chaired by Jim Snabe, and includes Stefan Heidenreich, Nigel Hinshelwood, Christophe Cuvillier and Tim Radford (joining Jan 2026). The group will meet periodically to guide Europe-first product development, policy engagement, and sustainability efforts, reinforcing Google Cloud’s commitment to regional expertise and customer-focused innovation.

🔐 Cloudflare unveiled a self‑serve BYOIP API enabling customers to onboard and manage their own IP prefixes via automated workflows. The new flow replaces manual LOA reviews with a two-step validation that uses RPKI ROAs plus either IRR route-object modification or a reverse DNS validation token. Cloudflare will auto-generate LOA-style documentation for operators that still require it and enforces a default service binding to prevent accidental prefix blackholing. The initial rollout supports prefixes originated from AS13335 and is designed to shorten deployment timelines while strengthening routing security.

🔒 Check Point expands CloudGuard to protect GenAI applications by extending the ML-driven, open-source CloudGuard WAF that learns from live traffic. The platform moves beyond traditional static WAFs to secure web interactions, APIs (REST, GraphQL) and model-integrated endpoints with continuous learning and high threat-prevention accuracy. This evolution targets modern attack surfaces introduced by generative AI workloads and APIs.

🚀 LMCache implements a node-local, tiered KV Cache on GKE to extend the GPU HBM-backed Key-Value store into CPU RAM and local SSD, increasing effective cache capacity and hit ratio. In benchmarks using Llama-3.3-70B-Instruct on an A3 mega instance (8×nvidia-h100-mega-80gb), configurations that added RAM and SSD reduced Time-to-First-Token and materially increased token throughput for long system prompts. The results demonstrate a practical approach to scale context windows while balancing cost and latency on GKE.

🤖 This recap of The Agent Factory features Logan Kilpatrick from Google DeepMind demonstrating vibe coding in Google AI Studio, a Build workflow that turns a natural-language app idea into a live prototype in under a minute. Live demos included a virtual food photographer, grounding with Google Maps, the AI Studio Gallery, and a speech-driven "Yap to App" pair programmer. The episode also surveyed agent ecosystem updates—Veo 3.1, Anthropic Skills, and Gemini improvements—and highlighted the shift from models to action-capable systems.