All news in category "Vendor and Hyperscaler Watch"

Visibility Gaps in Patching and Vulnerability Remediation

🔍 Modern patch management demands centralized visibility, faster prioritization, and accountable remediation to close growing exposure gaps. The article highlights how legacy systems such as WSUS and SCCM struggle with mixed environments, remote endpoints, and third-party applications, producing inconsistent patch states and unnoticed failures. Action1 is presented as a cloud-native platform that inventories endpoints, maps missing updates to CVEs, automates targeted deployments and retries failures, and provides audit-ready reporting to unify security and IT workflows.

Amazon S3 Adds Conditional Copy Support for Writes

🔐 Amazon S3 now supports conditional copy operations via the CopyObject API, enabling verification of an object's existence or content in the destination bucket before copying. You can supply the HTTP If-None-Match header to ensure the destination object does not exist, or If-Match with an ETag to validate content prior to copy. Administrators can enforce these checks using s3:if-match and s3:if-none-match bucket policy condition keys. This capability is available at no additional charge in all AWS Regions and removes the need for additional client-side coordination or pre-copy validation calls.

AWS Elastic Beanstalk: Corretto 25 with Tomcat 11 on AL2023

🚀 AWS Elastic Beanstalk now supports building and deploying Tomcat 11 applications using Amazon Corretto 25 on Amazon Linux 2023 (AL2023). The platform enables developers to leverage Java 25 and Jakarta EE 11 features such as compact object headers, ahead‑of‑time (AOT) caching, and structured concurrency while benefiting from AL2023’s security and performance improvements. Environments can be created through the Elastic Beanstalk Console, CLI, or API and are generally available in commercial and GovCloud regions.

Amazon EC2 High-Memory U7i-8tb Instances in London

🚀 AWS has launched Amazon EC2 U7i-8tb (u7i-8tb.112xlarge) instances in the Europe (London) region, offering 8 TiB of DDR5 memory and 448 vCPUs for memory-intensive workloads. Powered by custom fourth-generation Intel Xeon Scalable processors (Sapphire Rapids), these 7th-generation instances deliver up to 135% more compute than prior U-1 instances and support up to 100 Gbps for EBS and networking with ENA Express. They are aimed at mission-critical in-memory databases like SAP HANA, Oracle, and SQL Server.

Amazon EBS adds per-minute avg IOPS and throughput metrics

📈 Amazon EBS now emits two new per-volume CloudWatch metrics—VolumeAvgIOPS and VolumeAvgThroughput—providing one-minute average I/O and throughput visibility. These metrics are enabled by default at no extra charge for all EBS volumes attached to EC2 Nitro instances in Commercial Regions, including AWS GovCloud (US) and AWS China. Use them to monitor trends, troubleshoot performance bottlenecks, tune provisioned performance, and build dashboards or alarms to automate responses.

How We Escaped the Linux Networking Stack for Soft-Unicast

🐟 Cloudflare describes building "fish" (SLATFATF), a service to egress packets using soft-unicast address space and the challenges encountered with the Linux networking stack. They found that conntrack and Netfilter interactions can silently rewrite source ports and break connections, so they evaluated several approaches including Netlink manipulation, TCP_FASTOPEN_CONNECT sockets, and routing fixes. Ultimately they preferred terminating and proxying TCP locally to avoid fragile kernel workarounds, after testing that disabling early demux produced only modest CPU effects.

Building a High-Performance VPN with Linux for WARP

🛡️ Cloudflare explains how it initially implemented WARP as a Layer‑3 VPN by leveraging the Linux networking stack to egress arbitrary user packets from edge machines. They used a TUN device, nftables/Netfilter rules and the conntrack module to perform NAT, mark flows, and distinguish client traffic from locally‑originated traffic. Core tunnel handling was written in Rust (boringtun/WireGuard) and paired with MASQUE and defense‑in‑depth controls. The approach worked but required one IPv4 address per server, creating a scalability and cost challenge that led them to explore IP sharing.

Amazon Web Grounding for Nova Models Now Generally Available

🌐 Web Grounding is now generally available as a built-in tool for Nova models, usable today with Nova Premier via the Amazon Bedrock tool use API. It retrieves and incorporates publicly available information with citations to support responses, enabling a turnkey RAG solution that reduces hallucinations and improves accuracy. Cross-region inference makes the tool available in US East (N. Virginia), US East (Ohio), and US West (Oregon). Support for additional Nova models will follow.

Lower Your TCO with Check Point's SASE Platform

🔒 Check Point’s SASE offering and its TCO savings calculator explain how consolidating security and networking into a cloud-delivered SASE reduces operational complexity and costs compared with traditional VPN-based architectures. The article explains product, operational and performance savings and quantifies ROI improvements when replacing hardware-heavy VPN backhauling. It highlights benefits such as simplified management, faster time-to-value, improved throughput and reduced exposure to risk for distributed users.

Signal Rolls Out Quantum-Safe Triple Ratchet (SPQR)

🔐 Signal has rolled out a quantum-safe update that adds a third ratchet, branded SPQR, to its secure messaging protocol. Instead of replacing the existing Double Ratchet, Signal runs a parallel KEM-based ratchet and derives encryption keys by mixing outputs from both ratchets with a KDF. Developed with PQShield, AIST, and NYU, the design preserves familiar behaviors while adding post-quantum resilience for forward secrecy and post-compromise protection.

Google Public Sector Summit: A New Era for Government AI

🔔 At the Google Public Sector Summit in Washington D.C., leaders highlighted a shift toward agentic AI and large-scale cloud modernization. Google introduced Gemini for Government, an accredited platform providing an AI Agent Gallery, agent-to-agent protocols, enterprise connectors, and governance controls to deploy and monitor AI agents. Speakers showcased real-world deployments across defense, city, and education sectors, and Google announced expanded partner investments plus an enhanced partnership with NVIDIA to support on-premises and air-gapped environments.

Google Cloud launches unified home for technical docs

📚 Google Cloud has consolidated all technical documentation onto a new, dedicated platform to improve discoverability and support AI-driven experiences. By centralizing content on a unified site and integrating Gemini into authoring tools, Google aims to accelerate content creation and deliver context-aware assistance. The site offers faster performance, AI-powered translation across 12 languages, and preserves existing URL patterns to minimize disruption.

AWS Control Tower Now Available in Asia Pacific (NZ)

🚀 AWS Control Tower is now available in the AWS Asia Pacific (New Zealand) Region, bringing the service to 34 AWS Regions plus the AWS GovCloud (US) Regions. The service simplifies setup and governance of a secure, multi-account AWS environment, enabling a landing zone in 30 minutes or less and centralized visibility into compliance status. Existing customers can extend governance to the new region via the Control Tower settings by selecting regions and updating their landing zone; once applied, governed accounts, managed accounts, and registered organizational units (OUs) will be managed in the new region.

CrowdStrike Falcon Achieves 100% in SE Labs EPS Evaluation

🛡️ In SE Labs’ September 2025 Enterprise Endpoint Security evaluation, CrowdStrike Falcon earned the AAA EPS certification and recorded 100% Protection Accuracy, 100% Legitimate Accuracy and 100% Total Accuracy with zero false positives. SE Labs tested 75 targeted and 25 general attacks across full kill chains; Falcon detected and blocked or neutralized every attempt. The platform also won three SE Labs awards, including Enterprise Endpoint (Windows), Enterprise Ransomware, and Falcon Go for Small Business New Endpoint.

Giles AI on Google Cloud: Transforming Medical Research

🚀 Giles AI migrated its healthcare-focused platform to Google Cloud to reduce latency, improve scalability, and accelerate developer velocity. Using Google Kubernetes Engine, Cloud Run, and Compute Engine, the company orchestrates complex clinical data flows and routes prompts through Vertex AI and Model Garden to remain model-agnostic. Data storage and extraction are handled with Cloud SQL, Cloud Storage, and Document AI, while Cloud Armor and Security Command Center bolster security and compliance. Early customer results include dramatic reductions in research time and improvements in response accuracy.

Amazon ElastiCache Adds Dual-Stack IPv6 Service Endpoints

🌐 Amazon ElastiCache now provides dual-stack service endpoints, enabling management of resources over both IPv4 and IPv6. ElastiCache interface VPC endpoints powered by AWS PrivateLink also support dual-stack connectivity. The update, available in all AWS commercial, China, and GovCloud (US) Regions, helps simplify IPv6 migration and compliance without extra charges. This enables staged migrations and modernization while preserving existing IPv4 access.

Check Point's AI Cloud Protect with NVIDIA BlueField

🔒 Check Point has made AI Cloud Protect powered by NVIDIA BlueField available for enterprise deployment, offering DPU-accelerated security for cloud AI workloads. The solution aims to inspect and protect GenAI traffic and prompts to reduce data exposure risks while integrating with existing cloud environments. It targets prompt manipulation and infrastructure attacks at scale and is positioned for organizations building AI factories.

Microsoft and NVIDIA Deepen AI Infrastructure Partnership

🚀 Microsoft and NVIDIA announced expanded AI infrastructure on Azure, bringing NVIDIA RTX PRO 6000 Blackwell Server Edition to Azure Local, new Nemotron and Cosmos models via Azure AI Foundry, and broader support for Run:ai and GB300 NVL72 supercomputing clusters. These updates enable on-premises and edge AI with cloud-like management, improved GPU utilization, and infrastructure tailored for frontier reasoning, multimodal workloads, and real-time inferencing. Microsoft also highlighted NVIDIA Dynamo optimizations for ND GB200-v6 VMs to boost inference throughput at scale.

Google Cloud launches managed DRANET for GKE with A4X Max

🚀 Google Cloud is previewing managed DRANET on GKE, enabling Kubernetes to treat high-performance RDMA network interfaces as schedulable resources. The integration aligns NICs and GPUs by NUMA topology to reduce latency and increase throughput, while abstracting away operational complexity. It launches with the new A4X Max instances to deliver topology-aware networking for large multi-GPU AI workloads. Developers can request specific network interfaces in pod specs and rely on GKE to co-schedule NICs and accelerators, improving utilization and simplifying operations.

A4X Max, GKE Networking, and Vertex AI Training Now Shipping

🚀 Google Cloud is expanding its NVIDIA collaboration with the new A4X Max instances powered by NVIDIA GB300 NVL72, delivering 72 GPUs with high‑bandwidth NVLink and shared memory for demanding multimodal reasoning. GKE now supports DRANET for topology‑aware RDMA scheduling and integrates NVIDIA NeMo Guardrails into GKE Inference Gateway, while Vertex AI Model Garden will host NVIDIA Nemotron models. Vertex AI Training adds NeMo and NeMo‑RL recipes and a managed Slurm environment to accelerate large‑scale training and deployment.