All news in category “Vendor and Hyperscaler Watch”

🔁 This post explains a Python-based solution to migrate automation rules from Security Hub CSPM (ASFF) to the new Security Hub that adopts the open OCSF schema. The toolkit discovers rules across specified Regions, evaluates each rule against predefined ASFF→OCSF field mappings, and converts compatible rules into a CloudFormation template preserving order and Regional context. Actions or criteria without OCSF equivalents are flagged or partially migrated; migrated rules are created in a DISABLED state by default to allow review and testing. The package includes discovery, transformation, and template-generation scripts plus a migration report to guide manual adjustments.

🚀 Microsoft was named a Leader in the 2025 Gartner Magic Quadrant for AI Application Development Platforms and is positioned furthest for Completeness of Vision. The post presents Microsoft Foundry as a unified platform to build, deploy, and govern agentic AI—emphasizing secure grounding, multi-agent orchestration, observability, and cloud-to-edge model deployment. It also describes an agent-driven submission process that automated evidence collection and validation to improve accuracy and efficiency.

🚀 AWS announced the general availability of the new Amazon EC2 M8gn and M8gb instances powered by AWS Graviton4. M8gn targets network‑intensive workloads and provides up to 600 Gbps of networking bandwidth with 6th‑generation Nitro Cards, while M8gb focuses on high block‑storage throughput with up to 150 Gbps of EBS bandwidth. Both families offer up to 768 GiB of memory, large instance sizes, and EFA support on selected sizes; initial availability is in US East (N. Virginia) and US West (Oregon).

🚀 AWS Databases including Amazon Aurora PostgreSQL, Amazon Aurora DSQL, and Amazon DynamoDB are now generally available on the Vercel Marketplace, allowing developers to create and connect to managed databases directly from Vercel in seconds. You can provision a new AWS account from Vercel that includes access to these databases and $100 USD in credits usable for up to six months, and manage billing and usage via the AWS settings portal in the Vercel dashboard. Serverless options support scale-to-zero economics and are available in multiple Regions including US East, US West, Europe, and Asia Pacific.

🚀 Cluster Director is now generally available on Google Cloud, offering a managed, topology‑aware control plane for Slurm clusters to streamline AI and HPC operations. It automates Day 0 planning, Day 1 deployment, and Day 2 monitoring with health checks, Managed Lustre attachment, topology‑aware placement, and one‑click node remediation. There is no additional fee for Cluster Director — customers pay only for underlying Google Cloud resources. Google also announced a Preview of Slurm on GKE to combine Slurm's batch scheduling with Kubernetes' autoscaling and self‑healing.

📊 Administrators and support teams can now monitor the health and performance of Amazon WorkSpaces Applications fleets, sessions, instances, and users through a new set of Amazon CloudWatch metrics. These metrics can be enabled across fleets from the CloudWatch console and dynamically update to reflect current state, simplifying troubleshooting of end-user streaming sessions and performance investigations. To receive the metrics, fleets must use a WorkSpaces Applications image with the agent released on or after December 06, 2025, or be updated via Managed WorkSpaces Applications image updates released on or after December 05, 2025; metrics are available in AWS commercial and AWS GovCloud (US) Regions where the service operates.

🔐 The article introduces Access Fabric as a unified, adaptive approach that links identity, device, and network signals to make real-time, risk-based access decisions throughout every session. It warns that fragmented identity and network tools create visibility gaps that AI-empowered attackers can exploit. By being contextual, connected, and continuous, an Access Fabric closes seams, reduces complexity, and enforces consistent policies for users, devices, and AI agents.

🏆 Google Cloud was named a Leader in The Forrester Wave™: AI Infrastructure Solutions, Q4 2025, earning the top score for Current Offering and the highest possible marks in 16 of 19 evaluation criteria. The post emphasizes an integrated AI Hypercomputer that co‑designs custom TPUs with NVIDIA GPUs and a high‑bandwidth network fabric to optimize throughput, latency, and TCO. It highlights operational tools like GKE Autopilot and Cluster Director, cost controls such as Dynamic Workload Scheduler, and security investments including Titan chips and Security Command Center.

🚀 Amazon Aurora PostgreSQL-Compatible Edition now supports PostgreSQL 18.1 in the Amazon RDS Database Preview Environment, enabling evaluation of the new engine on Aurora without self-installation or manual provisioning. PostgreSQL 18.1, released September 9, 2025, adds skip scan for multicolumn B-tree indexes, improved WHERE handling for OR/IN, parallel GIN index builds, and join enhancements. Observability gains include buffer usage counts, index lookup details, and a per-connection I/O utilization metric. Preview clusters are retained up to 60 days and are priced the same as production Aurora instances in the US East (Ohio) Region.

📄 AWS Billing Conductor now supports service-specific custom line items scoped to a single AWS service or a selected set of services. Customers can control whether these line items appear itemized or consolidated in pro forma billing artifacts such as the Bills Page, Cost Explorer, and Cost and Usage Records. The enhancement enables more precise charge-back and re-billing workflows, including applying discounts to Saving Plans fees or allocating shared support charges. This capability is available now across AWS commercial Regions, excluding China.

⚡ Gemini 3 Flash expands the Gemini 3 family with a low-latency, cost-efficient model tuned for high-frequency enterprise workflows. It combines Pro-grade reasoning with Flash-level speed to enable near real-time multimodal processing, rapid agentic coding, and responsive interactive agents without sacrificing accuracy. Available in Gemini Enterprise, Vertex AI, and Gemini CLI, it targets scale and affordability for production deployments.

⚡Amazon EC2 Auto Scaling introduces the new LaunchInstances API, a synchronous operation that provides immediate feedback on capacity availability when provisioning instances inside an Auto Scaling group. Customers can specify placement overrides for any Availability Zone or subnet to control where capacity is launched. The call returns real-time insight so operators can implement alternative strategies if capacity is constrained, and it also supports optional asynchronous retries to help reach desired capacity. The feature is available in all AWS Regions and AWS GovCloud (US) Regions at no additional cost beyond standard EC2 and EBS usage.

🚀 Amazon EC2 C8g instances are now available in AWS Europe (Zurich). Powered by Graviton4 processors, C8g delivers up to 30% better performance versus Graviton3 and offers up to 3x more vCPUs and memory than prior C7g sizes. Built on the AWS Nitro System, these instances provide enhanced performance, security, and networking (up to 50 Gbps) for compute‑intensive workloads. Twelve sizes, including two bare‑metal options, support HPC, gaming, video encoding, CPU inference, and analytics.

🚀 AWS has expanded availability of Graviton4-based Amazon EC2 C8g, M8g, and R8g instances to AWS GovCloud (US-West and US-East). These instances run on AWS Graviton4 processors and the AWS Nitro System, delivering up to 30% better performance versus Graviton3 and larger instance sizes with up to 3x vCPUs and memory. Enhanced networking and EBS bandwidth support high-throughput workloads.

🔒 Amazon Elastic Container Registry (ECR) Public now supports PrivateLink for the US East (N. Virginia) SDK endpoint. This enables private, secure connectivity from an Amazon VPC to the ECR Public SDK endpoint, reducing exposure to the public internet. Organizations can maintain network privacy, meet stricter security requirements, and simplify access patterns when creating and managing ECR Public repositories. Use cases include protecting sensitive image pulls and streamlining CI/CD network architecture.

🔒 Sponsored content from Token presents wireless biometric passwordless authentication as a way to transform MFA from a persistent cost center into a measurable productivity gain. By replacing passwords and authenticator apps with proximity-bound biometric hardware such as Token Ring and Token BioStick, Token says average login time falls from 22 seconds to 2 seconds. The vendor asserts this yields roughly $1,466.67 per employee per year in recovered productivity while also reducing password resets and blocking phishing, session relay, and social-engineering attacks.

🚀 Amazon Web Services has introduced OI2 instances for Amazon OpenSearch Service, expanding the OpenSearch Optimized Instance family with sizes from large to 24xlarge and up to 22.5 TB of storage. The OI2 instances use the same architecture as OR2, pairing compute and 3rd-generation AWS Nitro SSDs for caching with remote S3-based managed storage, delivering up to 9% higher indexing throughput versus OR2 and up to 33% versus I8g in AWS internal benchmarks. OI2 is offered with pay-as-you-go and reserved pricing and is available in 12 AWS regions globally.

🔒 Palo Alto Networks sets a partner-first agenda for 2026 focused on integration, embedded AI and ecosystem-driven growth. The company encourages partners to consolidate point tools onto a unified platform, embed AI into core workflows and build repeatable services across consulting, resale and managed security. Program updates will simplify engagement, expand API and alliance integrations, and aim to increase partner profitability while accelerating customer time‑to‑value.

🔍 Modern SOCs frequently operate in a reactive mode, discovering threats only after incidents escalate. ANY.RUN's Threat Intelligence Lookup augments alerts with behavioral insight, infrastructure links, and sandbox observations so analysts can prioritize high-risk findings. Paired with continuous TI Feeds and industry/geographic attribution, teams reduce noise, speed triage, and tune detections to protect the business proactively.

⚠️ Deutsche Telekom has introduced Call Check, an automated warning feature that flags incoming calls listed in a database as suspicious or fraudulent. When a call from a domestic or foreign number is identified, the recipient's smartphone displays a Caution, possible fraud! message to warn the user. The system is applied automatically to customers on the Telekom network and joins similar protections already deployed by competitors such as Vodafone, while O2 has yet to implement an equivalent service.