All news in category “Vendor and Hyperscaler Watch”

⚡ Google Cloud has announced general availability of Lightning Engine for Managed Service for Apache Spark, available in both serverless and managed cluster modes. Built on Gluten and Velox with Google-engineered enhancements, the engine compiles Spark physical plans into native C++ to accelerate vectorized operations, window functions, and sorting. It also optimizes connectors for Cloud Storage and BigQuery and adds cost-based optimizations like hash table caching, aggregation pushdown, and adaptive shuffle partitioning.

🔍 Amazon VPC Flow Logs now supports EC2 resource tag embedding and next-hop interface metadata to simplify network monitoring and troubleshooting. With tag support you can include values from network interfaces, EC2 instances, and Auto Scaling groups, removing the need to join log data with external tag metadata. Next-hop metadata captures interface ID, subnet, AZ, VPC, and interface type to clarify traffic paths through NAT Gateways, NLBs, and Transit Gateways.

🚀 Amazon EC2 M9g and M9gd instances powered by AWS Graviton5 processors are now generally available. M9g targets general-purpose workloads and agentic AI tasks, while M9gd adds local NVMe SSD storage for low-latency, high-throughput use cases. These instances offer up to 25% better compute performance over Graviton4-based M8g/M8gd and deliver sizable gains for databases, web apps, and ML. They run on the sixth-generation Nitro System and introduce the Nitro Isolation Engine for formally verified workload isolation.

🚀 Amazon EC2 P6-B200 instances powered by NVIDIA Blackwell GPUs are now available in the AWS GovCloud (US-East) Region. These instances deliver up to 2x performance versus P5en for AI training and inference, with 8 GPUs and 1440 GB of high-bandwidth GPU memory. They include 5th Gen Intel Xeon processors, a 60% boost in GPU memory bandwidth, up to 3.2 Tbps EFAv4 networking, and run on the AWS Nitro System for secure scaling.

📣 AWS now allows customers to update AWS Cost and Usage Report 2.0 table configurations through the AWS Management Console and SDK/CLI. This change lets users modify export content, time granularity, column selection, export format, and destination settings without deleting and recreating exports. Updated preferences take effect starting with the next scheduled export delivery. The feature removes the previous requirement to recreate exports to adopt new CUR 2.0 schema enhancements.

🛡️ Today Cloudflare launched Application Services for Private Origins in closed beta for eligible Enterprise customers, enabling secure routing to private IP origins without exposing them to the public Internet. This lets Cloudflare’s WAF, bot management, rate limiting, Workers, and other services sit in front of private applications using existing private connectivity such as Cloudflare Tunnel, Cloudflare Mesh, or Cloudflare WAN. The feature uses a toggle on proxied DNS records or an API attribute to instruct Cloudflare’s private networking layer to route traffic to private networks, and extends Layer 4 support via Spectrum for TCP/UDP services.

🛡️ This article outlines how enterprises can detect and block unauthorized AI tools—ranging from public chatbots like ChatGPT and Claude to meeting recorders and local model runners. It recommends monitoring NGFW/web-filter logs, EDR/EPP and MDM tools, browser policies, DNS reroutes, and application allowlists. The guidance covers detection indicators (domains, executables, SNI, calendar invites) and concrete lockdown steps (category blocks, policy toggles, OAuth restrictions). Emphasis is placed on offering approved alternatives and using layered controls rather than outright bans.

🔐 The OMB Memo M-26-14 ends compliance-driven data hoarding and mandates a risk-based, machine-speed logging approach. It requires six months of logs to be hot and searchable and one year retrievable, expands visibility to IoT/OT, and emphasizes continuous event monitoring and AI-driven detection. Legacy SIEMs struggle to meet these goals; Palo Alto Networks positions Cortex XSIAM as a FedRAMP-certified solution built for index-free, AI-enabled, cross-domain threat hunting.

🚀 Amazon FSx for OpenZFS now supports the Intelligent‑Tiering storage class in eight additional AWS Regions across the US, Europe, Asia Pacific, and South America. The Intelligent‑Tiering class automatically moves data among Frequent Access, Infrequent Access, and Archive tiers based on usage patterns and can include an optional SSD read cache for active data. This provides high performance for active workloads and lower-cost storage for less-active data, with up to 85% savings versus FSx SSD and up to 20% versus on‑premises HDD NAS.

🚀 Amazon SageMaker Unified Studio Notebooks now support Amazon EMR Serverless with Apache Spark Connect, enabling PySpark and Spark SQL execution on EMR Serverless Spark Applications directly from notebook cells. Users can choose their Spark runtime from the Notebook side panel, with the selection applying to both Python and SQL cells. The feature integrates with SageMaker Data Agent to generate code and execution plans from natural language prompts and offers pre-initialized capacity, unified Spark UI monitoring, and VPC connectivity for isolated workloads. It is available in all regions where SageMaker Unified Studio is offered and supports both Unified Studio notebooks and JupyterLab IDEs.

🔍 Microsoft outlines a structured approach to investigate AI interactions across Microsoft 365 Copilot and Azure AI services, emphasizing telemetry from Purview, Defender, and Sentinel. The new investigator playbook follows a scope–context–signal methodology to identify who interacted with AI systems, what resources were accessed, and when events occurred. It operationalizes detection logic, KQL queries, and schema references to help response teams build coherent investigative narratives and assess impact.

🛈 Amazon S3 Access Grants are now available in the AWS European Sovereign Cloud (Germany) Region. The feature maps identities from directories like Microsoft Entra ID and AWS IAM principals to S3 datasets, enabling automated, scalable data permission management. This simplifies granting S3 access to end users based on corporate identities. Check the AWS Region Table for full regional availability and refer to the product page for details.

🔒 Meta announced it will repurpose information businesses share about users' activity off its platforms to personalize Feed content and AI chatbot responses, expanding beyond targeted ads. The company said no new data collection is involved and that users can control this through an updated "Activity from other businesses" setting, replacing "Your activity off Meta technologies." The change will roll out next month in the U.S. and several other countries.

🚀 Claude Fable 5 is now available in Microsoft Foundry, powering agents across GitHub Copilot and the Foundry Agent Service to tackle long-running, multi-stage tasks such as complex refactoring, research synthesis, and document-heavy workflows. Foundry adds enterprise-grade security, governance, and operational controls to help organizations evaluate, deploy, and scale autonomous systems in production. Anthropic and Microsoft combine safeguards, guided guardrails, and observability to support responsible use while enabling powerful multimodal reasoning and continuous agent improvement.

🤖 Microsoft has integrated Anthropic’s Claude Fable 5 into Foundry, bringing Mythos-level capabilities to GitHub Copilot and Foundry Agent Service with enterprise-grade safeguards. The model excels at long-running, multi-stage tasks—code refactors, deep research, and document-heavy workflows—while Foundry adds governance, observability, and deployment controls. Combined with Microsoft IQ, Fable 5 can reason across organizational data and applications to support production-grade autonomous agents.

🛡️ Google details how Google Security Operations pairs with Google AI Threat Defense to detect, investigate, and contain AI-accelerated attacks across cloud and enterprise environments. The post explains three specialized agents — Detection Engineering, Triage and Investigation, and Threat Hunting — that translate threat intelligence into custom detections, autonomously investigate alerts, and proactively hunt stealthy compromises. These agents use diverse telemetry, simulated events, and AI-driven automation to reduce time-to-detection and speed remediation, addressing gaps where patching is impossible or delayed.

🚀 GKE Inference Gateway uses prefix caching and model-aware routing to reduce accelerator idle time and speed up LLM inference. By matching request prefixes to pods that already hold the KV cache, it avoids repeated recomputation and lowers latency compared with naive round-robin load balancing. Independent benchmarks show 15.7% higher throughput, 92.8% faster time-to-first-token, and 62.6% lower inter-token latency. Snap reports 75–80% prefix cache hit rates in production integrations.

🔍 Storage Insights datasets now include activity insights that provide near-real-time visibility into object and bucket operations across your Google Cloud Storage estate. These BigQuery-linked views expose object-level writes, updates, deletes and errors, bucket-level aggregates and regional traffic patterns to support cost optimization and faster troubleshooting. The feature is generally available and customizable by org, folder, project, or specific buckets, enabling queries, Looker visualizations, and integration with other Storage Intelligence capabilities.

🔍 Today AWS announced the preview of AWS FinOps Agent, a frontier agent designed for FinOps practitioners and engineering teams to answer cost questions, surface optimization opportunities, and automate investigations of cost anomalies. The agent can generate cloud cost reports, surface rightsizing and Savings Plans recommendations from AWS Cost Optimization Hub and AWS Compute Optimizer, and create Jira tickets. It can post anomaly findings to Slack and run recurring FinOps workflows on a defined schedule. The preview is available in US East (N. Virginia) and covers cost and usage data from most Regions; it is offered at no additional charge during the preview.

🚀 AWS has launched general availability for metal-48xl and metal-96xl sizes across several M8 and R8 EC2 instance families powered by custom sixth-generation Intel Xeon Scalable processors and the latest AWS Nitro cards. These new sizes offer up to 43% better compute performance per vCPU over prior generations and deliver up to 600 Gbps network bandwidth or up to 300 Gbps EBS bandwidth depending on the variant. Selected sizes support Elastic Fabric Adapter (EFA) for lower latency in tightly coupled clusters and the metal sizes are initially available in the AWS US East (N. Virginia) region.