All news in category “Vendor and Hyperscaler Watch”

🔍 AWS IoT Core now emits two new Amazon CloudWatch Log event types to help troubleshoot device connectivity and authentication across IoT fleets. The Ping log captures MQTT Keep‑alive messages to identify connections or devices that fail to maintain connectivity. The Connection.AuthNError log records rejected connection attempts with detailed error codes to speed resolution of credential and certificate issues. Enable event-level logging and choose a CloudWatch log group, then opt into these event types; they are available in all AWS Regions where AWS IoT Core operates.

🔔 Amazon Connect Customer now delivers real-time dashboard alerts when specified keywords, phrases, or sentiment patterns arise during live calls and chats. Supervisors can listen to live conversations, view real-time transcripts and sentiment analysis, and coach agents via chat to address issues such as cancellation requests before customers churn. The feature is available in all AWS commercial and AWS GovCloud (US-West) regions offering conversational analytics.

🐧 Microsoft unveiled Coreutils for Windows at Build 2026, delivering many familiar Linux command-line utilities as native Windows applications. Based on the open-source uutils project (a Rust rewrite of GNU coreutils), the release aims to let developers use the same commands across Linux, macOS, Windows, and WSL without changing workflows. The package is available on GitHub and via WinGet as a single coreutils.exe binary that exposes individual command names through NTFS hardlinks. Microsoft notes some command conflicts, POSIX limitations, and behavioral differences on Windows.

🧭 OpenAI has updated the GPT-5.5 Instant model to improve answer accuracy, pacing, and conversational style while reducing long, bullet-heavy responses to sound more natural. The company will retire legacy models: o3 on August 26 with a 90-day sunset and GPT-4.5 on June 27 with a 30-day sunset. Additionally, OpenAI is integrating a job search tool into ChatGPT to surface live listings and help tailor resumes, and it has enhanced resume editing and export capabilities. These changes are rolling out globally to paid users.

🔔 AWS now supports Kubernetes version 1.36 in Amazon EKS and Amazon EKS Distro. You can create new clusters or upgrade existing ones using the EKS console, eksctl, or infrastructure-as-code tools across all AWS Regions, including GovCloud. Key features in 1.36 include GA User Namespaces, Mutating Admission Policies for CEL, In-Place Pod-Level Vertical Scaling, and Resource Health Status reporting. EKS Distro images are available in ECR Public Gallery and GitHub, with documentation covering upgrade guidance and lifecycle policies.

🧭 This blog documents an experiment using Google Cloud to deploy a Gemma 3 inference workload across two regional GKE clusters, leveraging TPU v6e instances, managed DRANET for accelerator networking, and a multi-cluster Inference Gateway for cross‑region routing and failover. It describes building VPCs, reserving internal IPs, configuring Cloud Storage FUSE for model storage, creating TPU node pools with managed DRANET, registering clusters into a GKE Fleet, and deploying the inference server and gateway with health checks and autoscaling metrics. The objective is resilient, low‑latency routing to the nearest region with automatic failover to the other region if one fails.

🧭 Microsoft outlines a shift from isolated AI tools to a unified, enterprise-grade agent platform that runs real work. The post emphasizes a single integrated system spanning Azure, GitHub, Microsoft IQ, Foundry, Agent 365, and Microsoft 365 to build, contextualize, run, govern, and improve agents. It stresses secure-by-design governance, model choice, continuous improvement through feedback and tuning, and production-grade runtimes. The approach centers developers and enterprise context to make agents trustworthy and scalable.

🔐 AWS KMS introduced the GetKeyLastUsage API to report the date, time, operation type, CloudTrail event ID, and KMS request ID for the most recent cryptographic operation on a key. The feature works for customer-managed and AWS-managed keys across specs and origins and is visible in the AWS Management Console and AWS CLI. Tracking of last usage began on April 23, 2026 for most Regions, so historical gaps before tracking began should be considered. Use DisableKey, monitoring, and the kms:TrailingDaysWithoutKeyUsage condition to prevent accidental deletions.

🔒 AWS Config now supports internal service linked rules, allowing AWS services to evaluate resource configurations using AWS Config managed rules. These rules let AWS services like AWS Security Hub CSPM deploy and manage service-specific evaluations, with results sent directly to the deploying service. Evaluations occur at no charge from AWS Config and run independently of customer-managed recorders and rules, preserving existing inventory and compliance workflows. The feature is available in commercial, GovCloud, and China Regions.

🔒 At Microsoft Build 2026, Microsoft announced new security capabilities to integrate protection across the development lifecycle, addressing insecure code, agent proliferation, and model risk. The expanded preview of the multi-model agentic scanning harness (codename MDASH) integrates with Microsoft Defender to orchestrate hundreds of AI agents for exploit discovery. New tools such as Agent 365, MXC SDK, and Purview enhancements provide runtime controls, data protection, and governance to help developers and security teams act earlier and with consistent oversight.

🚀 Google Cloud Storage (GCS) is positioned as the preferred home for large-scale unstructured data and as a foundational component for production AI agents. This post highlights customer examples—Palo Alto Networks and Snap—using GCS as agent memory and analytics storage, and explains how the Model Context Protocol (MCP) enables secure, standardized access. Google offers two MCP server options: a fully managed Remote MCP server for easy, scalable deployments, and a self-managed Local MCP server for custom tooling and transformations, both integrated with Google Cloud security, observability, and tooling.

🔧 AWS Deadline Cloud now supports persistent storage for Service-Managed Fleets (SMF), enabling data retention across worker lifecycle events. Previously, SMF workers used ephemeral storage and required reinstalling software and assets after recycling. Now Deadline Cloud attaches persistent Amazon EBS volumes to SMF workers to preserve Conda environments, Perforce workspaces, shader caches, and asset collections, reducing startup time. You can configure volume counts per worker and set TTLs to balance retention and cost; availability and pricing match existing SMF EBS offerings.

🧭 Microsoft highlights how AI-driven agentic workflows demand a shared data context. Microsoft Fabric is presented as a unified data and AI platform that enables developers and agents to build production-ready apps by providing consistent organizational context. New announcements include the open-source Rayfin SDK/CLI for rapid backend deployment and Azure HorizonDB (PostgreSQL-compatible) in public preview, optimized for AI workloads.

🚀 Microsoft announced early access preview of Azure Cobalt 200 Arm-based VMs at Build 2026, delivering up to 50% generational performance improvements over Cobalt 100 for agentic AI and cloud-native Linux workloads. The Cobalt 200 SoC, built on Arm Neoverse V3 and TSMC 3nm, features chiplets, custom accelerators, and enhanced memory and security capabilities. New VM families (including high-memory and dense local storage) expand deployment choices and are available in selected preview regions.

🔧 Amazon SageMaker Studio's quick setup now completes in under twenty seconds, down from over two minutes, letting users rapidly move from sign-in to a fully configured Studio environment. Newly created Studio environments automatically receive serverless model customization permissions via a new managed policy, AmazonSageMakerModelCustomizationCoreAccess, enabling fine-tuning, evaluation, and deployment without manual IAM role configuration. Existing environments receive actionable guidance to add the permissions. The feature is available in all AWS Commercial Regions that support SageMaker Studio.

🔒 This post shows how SaaS providers can use Amazon Bedrock AgentCore resource-based policies to control multi-tenant access to a shared AgentCore Runtime and Runtime endpoint. It walks through two tenant scenarios: cross-account access for Example Corp and VPC-restricted access for AnyCompany, demonstrating how to apply resource-level Allow and explicit Deny conditions. The article covers required IAM permissions, example policy files, and verification steps to ensure network- and identity-based constraints are enforced.

🔎 Foundry IQ streamlines bringing enterprise and external knowledge into agent workflows by unifying content, improving ingestion, and offering a serverless model for retrieval. The service provides a Model Context Protocol (MCP) server, integrates Microsoft Web IQ for low-latency external context, and includes GA security and compliance features. Serverless Developer tier is in public preview with CU-based billing estimates and scale-to-zero capacity.

🛠️ Microsoft Foundry presents a unified platform to select, evaluate, optimize, and operate AI models across the full application lifecycle. The post emphasizes that production systems require continuous model selection, validation on real data, cost and latency management, and governance rather than simply picking the most capable model. Foundry adds new model families and Fireworks AI for production-grade open model inference via a single Azure endpoint with enterprise SLAs. It provides model routing, benchmarking with custom datasets, continuous evaluation, and operational controls like versioning, observability, and rollout strategies.

🔍 At Google Cloud Next, Google announced the preview of graph algorithms in Spanner Graph, bringing Google Research’s graph mining capabilities natively to Spanner to deliver faster, cost-effective, large-scale structural analytics. The feature integrates with ISO GQL, runs on dedicated compute to avoid transactional impact, and scales to graphs with tens of billions of edges. Results can be written back to Spanner or exported to Cloud Storage.

🚀 The post announces gcs-analytics-core, an open-source Java library that centralizes performance optimizations for Google Cloud Storage (GCS) across analytics engines like Apache Iceberg and Spark. Integrated natively in Apache Iceberg 1.11.0+, the library provides vectored I/O and smart Parquet footer prefetching to reduce I/O latency and improve throughput. Benchmarks using TPC-DS demonstrate sizable scan and execution time improvements across dataset sizes, and the project is available on GitHub for contributions and review.