All news in category “Vendor and Hyperscaler Watch”

🔧 Cloudflare’s Gen 13 platform pairs its Rust-based FL2 software rewrite with a redesigned server built around the 192-core AMD EPYC 9965, 768 GB DDR5-6400, PCIe 5.0 NVMe storage, and dual 100 GbE networking. The new configuration targets up to 2x throughput versus Gen 12 while improving performance-per-watt by up to 50% and increasing per-rack throughput. Gen 13 also adds PCIe encryption, expanded NVMe capacity and front-bay U.2 options, and support for two double-width PCIe accelerators to future-proof deployments.

🚀Cloudflare announces Gen 13 servers based on AMD EPYC™ 5th Gen Turin and the new Rust-based FL2 request layer. Turin delivers up to 192 cores and improved IPC but reduces per-core L3 cache; FL1's cache-heavy design produced unacceptable latency at high utilization. By rewriting the serving stack, Cloudflare unlocks up to 2x throughput over Gen 12 and up to 50% better performance-per-watt while keeping latency within SLAs.

🔐 Prisma Browser for Business delivers a secure, browser-first workspace tailored to small businesses, combining enterprise-grade threat detection with a simplified admin experience. Built on Chromium and powered by Palo Alto Networks' Precision AI, it inspects webpages and extensions in real time to block AI-enabled phishing, browser-delivered malware, and risky extensions. The browser adds AI-aware controls to prevent sensitive data from being pasted into unauthorized GenAI tools and offers one-click SSO onboarding, preconfigured policies, and a 30-day no-contract trial.

⚠️ Microsoft is addressing an intermittent Exchange Online outage that has blocked some users from accessing mailboxes via Outlook mobile and the new Outlook for Mac client since Thursday. The company traced the issue (incident EX1256020) to a newly introduced virtual account and is reverting that change across affected environments. Restarts failed to resolve the problem, and Microsoft is working on a permanent remediation and will provide a timeline when available.

🏥 AWS HealthImaging is now available in the AWS Europe (London) Region, offering a HIPAA-eligible, fully managed service to store, analyze, and share medical images at petabyte scale. The service provides DICOMWeb APIs for integration with existing clinical systems and AWS-native APIs for cloud-first implementations. AWS states organizations can reduce storage costs by up to 40%, enable faster image access for clinical workflows, and accelerate development of AI-powered diagnostic applications while maintaining strict security controls.

🛡️ CrowdStrike is extending Falcon Next‑Gen SIEM to ingest and operationalize telemetry from third‑party EDRs, beginning with Microsoft Defender, without requiring a Falcon sensor. The release embeds real‑time data pipelines via Falcon Onum to filter, enrich, and route telemetry, and expands federated search to include Falcon LogScale, ExtraHop, and cloud archives. It also introduces Third‑Party Indicator Management to operationalize external threat intelligence and a Query Translation Agent to convert legacy searches into CQL. Together these capabilities aim to reduce ingestion costs, accelerate investigations, simplify SIEM migrations, and let teams modernize SOC operations without replacing endpoint agents.

🔒 Google is introducing Advanced Flow, a new Android mechanism that lets power users sideload APKs from unverified developers while adding multi-step protections. The one-time process requires enabling Developer Mode, confirming you are not being coached by a threat actor, restarting and reauthenticating, then waiting one day to validate the changes. After completion users may enable installations for a week or indefinitely, and Android will display a warning that the app is from an unverified developer. The flow is intended to add friction and disrupt urgency-driven scam tactics.

🔊 Amazon Bedrock AgentCore Runtime now supports WebRTC for low-latency, bidirectional streaming between clients and agents, enabling real-time audio and video in browser and mobile applications. WebRTC complements existing WebSocket support by providing peer-to-peer, UDP-based media transport optimized for voice agents and other media-intensive experiences. AgentCore Runtime supports managed TURN via Amazon Kinesis Video Streams, third-party providers, or self-hosted TURN, and the capability is available in 14 AWS Regions.

🔒 Amazon EKS now offers a 99.99% Service Level Agreement for clusters running on the Provisioned Control Plane, up from the 99.95% SLA on the standard control plane. The upgraded SLA is measured in 1-minute intervals to deliver a more granular availability commitment for mission-critical workloads. At the same time, EKS introduces an 8XL scaling tier that doubles Kubernetes API server request processing capacity compared with the 4XL tier. Both the new SLA and the 8XL tier are available today in all regions where the Provisioned Control Plane is offered.

🔧 AWS announced the Neuron Dynamic Resource Allocation (DRA) driver for Amazon EKS, enabling Kubernetes-native, hardware-aware scheduling on Trainium-based instances. The driver publishes detailed device attributes — including hardware topology and Neuron-EFA PCIe co-location — directly to the Kubernetes scheduler, removing the need for custom scheduler extensions. Infrastructure teams can publish reusable ResourceClaimTemplates, while ML engineers reference them to deploy workloads without manual hardware tuning.

🔊 Amazon Web Services announced general availability of 10 new highly expressive Generative voices in Amazon Polly, covering eight locales including American, British, New Zealand, and Singapore English, plus French, Italian, German, and Swiss German. The Generative engine is now hosted in two additional regions — Europe (London) and Canada (Central) — and introduces a Bidirectional Streaming API. The new streaming capability lets customers send text and receive synthesized audio simultaneously, simplifying low-latency integrations with LLM-based systems for chatbots, game characters, and other real-time speech applications.

🔒 Microsoft presents an end-to-end strategy to secure agentic AI with the new Agent 365 control plane and updates across Microsoft Defender, Entra, Purview, and Sentinel. Announced for RSAC 2026, these measures focus on visibility, continuous identity protection, data loss prevention for Copilot prompts, and prompt-injection defenses to help organizations observe, govern, and defend agent ecosystems at scale.

📢 AWS Database Migration Service (DMS) Schema Conversion with GenAI is now available in nine additional AWS Regions, enabling local processing and helping meet data residency requirements. The feature leverages Amazon Bedrock foundation models — including Claude 3.5 Sonnet v2, Claude 3.7 Sonnet, and Claude Sonnet 4 — to automate schema and code conversion. It converts schemas and code from Oracle, SQL Server, MySQL, PostgreSQL, and Sybase to Amazon Aurora PostgreSQL-Compatible Edition and Amazon RDS for PostgreSQL, reducing manual effort and accelerating migrations. DMS Schema Conversion is available at no additional charge and can be accessed through the AWS Management Console or CLI.

🔐Google announced a new advanced flow for Android sideloading that imposes a mandatory 24-hour wait and biometric or PIN confirmation before permitting installs from unverified developers. The measure complements a developer verification mandate and is intended to make social‑engineering and rapid coercion attacks harder. Google will also offer free limited distribution accounts for hobbyists and students and says the flow does not apply to ADB installs; the changes roll out in August 2026 ahead of verification rules.

🔐 AWS DataSync now integrates with AWS Secrets Manager for credential management across all DataSync location types, including HDFS and Amazon FSx variants. Customers can centralize secrets in their account and optionally encrypt them with a customer-managed AWS KMS key to meet governance requirements. DataSync supports providing a secret ARN you manage or having DataSync automatically create and manage secrets. This capability is available in the majority of AWS regions where DataSync is offered.

🛡️ AWS Firewall Manager is now available in the AWS Asia Pacific (New Zealand) Region. The service centralizes policy management so cloud security administrators and site reliability engineers can protect applications while reducing the operational overhead of manually configuring and maintaining rules. With AWS Firewall Manager, customers can enforce defense-in-depth policies across AWS security services and create and manage AWS WAF security policies at scale. See the product documentation and region table for detailed availability and setup guidance.

📈 The AWS MCP Server (Preview) now publishes operational metrics to Amazon CloudWatch and adds semantic similarity search for Agent SOP discovery. Metric publishing under the AWS-MCP namespace tracks invocation counts, success rates, client and server errors, and throttling for tools such as call_aws and retrieve_agent_sop, enabling alarms and visibility into agent-driven activity. The documentation search (search_documentation) now returns relevant Agent SOPs alongside AWS docs, helping AI assistants find the correct workflows.

🔁 AWS now allows customers to use EC2 Fleet to consume interruptible Capacity Reservations by specifying reservation IDs across Launch Templates in a single Fleet call. When On‑Demand Capacity Reservations are idle, owners can make them temporarily available to accounts within an AWS Organization as interruptible reservations. The capability is available in all AWS commercial regions and aims to improve utilization and lower costs.

🤖 Cloudflare’s Workers AI now hosts frontier open-source models, beginning with Kimi K2.5, a 256k-context model that supports multi-turn tool calling, vision inputs, and structured outputs. The release enables organizations to run full agent lifecycles on Cloudflare’s Developer Platform, leveraging primitives like Durable Objects and Workflows. Cloudflare emphasizes improved price-performance, prefix caching, a session-affinity header, and a redesigned asynchronous API to lower latency and inference costs for agentic workloads.

🔒 Microsoft announced Zero Trust for AI, extending proven Zero Trust principles across the AI lifecycle and shipping new tools and guidance to help security teams deploy AI with confidence. The update adds an AI pillar to the Zero Trust Workshop, expands the Zero Trust Assessment to include Data and Networking, and introduces a Zero Trust for AI reference architecture. Microsoft also published practical patterns for threat modeling and AI observability to help teams verify agents, apply least privilege, and assume breach.