AWS
(29 articles)Database Savings Plans now supports Amazon OpenSearch Service and Amazon Neptune Analytics
AWS expands Database Savings Plans to include Amazon OpenSearch Service and Amazon Neptune Analytics, offering up to 35% savings for a one-year commitment. This helps cloud engineers optimize costs for these database services.
AWS Shield network security director findings are now available in AWS Security Hub
AWS Shield network security director findings, currently in preview, are now integrated with AWS Security Hub. This enhances cloud security posture by centralizing visibility into missing or misconfigured network security services.
Introducing Amazon Connect Health, Agentic AI Built for Healthcare
Amazon Connect Health is now generally available, providing purpose-built agentic AI for healthcare organizations to streamline patient engagement and point-of-care workflows. This is a significant new AI service launch for a specific industry vertical.
AWS HealthLake announces data transformation agent for automated CCDA-to-FHIR data conversion (Preview)
AWS HealthLake introduces a data transformation agent (preview) to automate legacy clinical document conversion to FHIR resources in days. This accelerates data interoperability and analytics for healthcare organizations.
Amazon Lightsail now offers OpenClaw, a private self-hosted AI assistant
Amazon Lightsail now allows simple and secure deployment of OpenClaw, a private self-hosted AI assistant, on your cloud infrastructure. This new offering provides an accessible way to run autonomous private AI agents.
Amazon GameLift Servers launches DDoS Protection
Amazon GameLift Servers now includes DDoS Protection, a new feature designed to improve game session resiliency for multiplayer games. This is a critical security enhancement for game developers.
Policy in Amazon Bedrock AgentCore is now generally available
Policy in Amazon Bedrock AgentCore is now generally available, offering centralized, fine-grained controls for agent-tool interactions. This enhances security and compliance for AI agent deployments.
Multi-party approval now supports approval team baselining
Multi-party approval (MPA) now allows administrators to run test approvals to confirm team setup and approver activity. This improves operational reliability and security for critical approval workflows.
AWS simplifies IAM role creation and setup in service workflows
AWS Identity and Access Management (IAM) now simplifies IAM role creation and configuration directly within service workflows. This improves security best practices by making it easier to set up appropriate permissions.
AWS Config now supports 30 new resource types
AWS Config now supports 30 additional AWS resource types, including Amazon Bedrock AgentCore and Amazon Cognito. This expands coverage for discovering, assessing, auditing, and remediating resources, enhancing compliance and security.
Amazon Redshift introduces reusable templates for COPY operations
Amazon Redshift now supports reusable templates for the COPY command, simplifying data ingestion workflows. This is a useful best practice for data engineers working with Redshift.
Amazon Redshift introduces new array functions for semi-structured data processing
Amazon Redshift now supports nine new array functions for processing semi-structured data in the SUPER data type. This enhances data processing capabilities and performance for analytics workloads.
Amazon EC2 R8g instances now available in additional regions
Amazon EC2 R8g instances, powered by AWS Graviton4 processors, are now available in AWS Middle East (UAE), Mexico (Central), and Europe (Zurich) regions. This expands access to high-performance, cost-effective compute.
Amazon Redshift Serverless now maintains datashare permissions during restore
Amazon Redshift Serverless now preserves datashare permissions when restoring a snapshot to the same namespace. This simplifies data sharing workflows and reduces administrative overhead.
AWS Elastic Beanstalk now offers AI-powered environment analysis
AWS Elastic Beanstalk now provides AI-powered environment analysis to quickly identify root causes and recommend solutions for health issues. This enhances operational efficiency and troubleshooting for applications.
Amazon OpenSearch Service introduces capacity optimized blue/green deployments
Amazon OpenSearch Service now offers a Capacity Optimized option for blue/green deployments, ensuring domain updates complete even with less available instance capacity. This improves deployment reliability and efficiency.
Accelerate Lambda durable functions development with new Kiro power
AWS introduces the Lambda durable functions Kiro power, enabling faster development of resilient, long-running multi-step applications and AI workflows with AI agent-assisted development. This is a significant tool update for serverless and AI development.
Amazon OpenSearch Ingestion now supports Amazon Managed Service for Prometheus as a sink
Amazon OpenSearch Ingestion now supports Amazon Managed Service for Prometheus as a sink, enabling fully managed, end-to-end metrics ingestion pipelines. This simplifies observability setups for cloud engineers.
Amazon OpenSearch Ingestion now supports unified ingestion endpoint for OpenTelemetry data
Amazon OpenSearch Ingestion now offers a unified ingestion endpoint for all three OpenTelemetry observability signals (logs, metrics, traces) through a single pipeline. This simplifies OpenTelemetry data management.
Amazon EventBridge Scheduler now provides a higher default quota for the CreateSchedule API
Amazon EventBridge Scheduler has increased its default quota for the CreateSchedule API action to 5,000 requests per second in 11 AWS Regions. This improves scalability for event-driven architectures.
AWS Batch now supports configurable scale down delay
AWS Batch now allows configuring a scale down delay for managed compute environments, reducing job processing delays for intermittent workloads. This helps optimize costs and performance for batch jobs.
Amazon SageMaker Unified Studio adds light mode support for IAM-based domains
Amazon SageMaker Unified Studio now supports light mode for IAM-based domains, allowing users to choose their preferred visual interface. This is a user experience improvement for data scientists and ML engineers.
Amazon EC2 I8ge instances now generally available in Europe (Ireland) AWS region.
Amazon EC2 I8ge instances, powered by AWS Graviton4 processors, are now available in the Europe (Ireland) region. These instances offer up to 60% better compute performance for storage-optimized workloads.
Amazon EC2 M8g instances now available in additional regions
Amazon EC2 M8g instances, powered by AWS Graviton4 processors, are now available in Africa (Cape Town), Asia Pacific (Malaysia), Europe (Milan, Zurich), and Canada West (Calgary) regions. This expands access to high-performance general-purpose compute.
Amazon SageMaker HyperPod now provides comprehensive observability for Restricted Instance Groups
Amazon SageMaker HyperPod now offers comprehensive observability for Restricted Instance Groups, giving teams deep visibility into compute resources and training workloads. This is crucial for optimizing foundation model training.
Amazon SageMaker Unified Studio adds metadata sync with third-party catalogs
Amazon SageMaker Unified Studio now supports metadata and context sync with Atlan, Collibra, and Alation. This provides a consistent view of data and AI assets across different catalog platforms.
Amazon SageMaker Unified Studio now supports AWS Glue 5.1 for data processing jobs
Amazon SageMaker Unified Studio now supports AWS Glue 5.1 for Visual ETL, notebook, and code-based data processing jobs. This update brings Apache Spark 3.5.6, Python 3.11, and Scala 2.12.18 to data scientists.
Amazon SageMaker Unified Studio launches support for remote connection from Kiro IDE
AWS announces the ability to remotely connect from Kiro IDE to Amazon SageMaker Unified Studio. This allows data scientists and ML engineers to leverage their Kiro setup for AI agent-assisted development.
OpenSearch OR2 and OM2 instances in AWS GovCloud (US-East, US-West) Regions
Amazon OpenSearch Service expands availability of OR2 and OM2 OpenSearch Optimized Instance families to 12 additional regions, including AWS GovCloud. These instances offer significant indexing and indexing throughput improvements.
GCP
(15 articles)H4D VMs, now GA, deliver exceptional performance and scaling for HPC workloads
Google Cloud's H4D VMs are now generally available, offering exceptional performance and scalability for HPC workloads powered by 5th Gen AMD EPYC™ processors. This is a significant performance boost for compute-intensive tasks.
From network APIs to network AI agents — building the agentic future of connectivity with Nokia
Google Cloud announces integration with Nokia's Network as Code platform, moving towards an agentic ecosystem for automating telecommunications workflows. This highlights a significant industry trend in AI-driven network operations.
Proactive Preparation and Hardening Against Destructive Attacks: 2026 Edition
This GCP blog post provides guidance on proactive preparation and hardening against destructive attacks, leveraging threat intelligence insights. It's crucial for cloud security architects to understand and implement these best practices.
Make security simpler: Introducing the Google Cloud recommended security checklist
Google Cloud has published a recommended security checklist to help organizations manage security requirements and configurations. This is a valuable resource for establishing a secure foundation in GCP.
Look What You Made Us Patch: 2025 Zero-Days in Review
Google Threat Intelligence Group tracked 90 zero-day vulnerabilities exploited in-the-wild in 2025, offering insights into the evolving threat landscape. This is important for understanding current cloud security risks.
Coruna: The Mysterious Journey of a Powerful iOS Exploit Kit
Google Threat Intelligence Group identified "Coruna," a powerful iOS exploit kit targeting iPhones running iOS 13.0 to 17.2.1. This threat intelligence is vital for securing mobile access to cloud resources.
Grow your own way: Introducing native support for custom metrics in GKE
GKE now natively supports custom metrics for autoscaling workloads, allowing platform engineers to scale based on application-specific signals. This provides greater flexibility and cost efficiency for Kubernetes deployments.
The AI-native core: Highly resilient telco architecture using Google Kubernetes Engine
This article explores building highly resilient, AI-native core architectures for telecommunications using Google Kubernetes Engine. It's a key industry trend and best practice for modernizing telco infrastructure.
Scaling the autonomous network: Introducing the Data Steward and Core Network Agents
Google Cloud introduces Data Steward and Core Network Agents as part of its Autonomous Network Operations framework. These new agents help communication service providers transition to zero-touch operations.
Designing private network connectivity for RAG-capable gen AI apps
This article provides a reference architecture for secure and reliable private network connectivity for Retrieval-Augmented Generation (RAG) generative AI applications on Google Cloud. It's a crucial best practice for AI deployments.
Unified Maintenance: A new, unified way to manage maintenance across Google Cloud
Google Cloud introduces Unified Maintenance to centralize and simplify the management of planned maintenance across services. This improves operational efficiency and business continuity.
Fueling the autonomous network: Google Cloud and DigitalRoute simplify data readiness for AI
Google Cloud partners with DigitalRoute to simplify data readiness for AI in the telecommunications sector, enabling self-healing and self-optimizing networks. This addresses a critical challenge for AI adoption in telco.
The rise of the autonomous network: How GraphML is redefining telecom operations
This article discusses how GraphML and digital twins are enabling autonomous networks in telecommunications, managing complex multi-layer ecosystems. It highlights advanced AI/ML applications in cloud-native infrastructure.
Data Strategy = AI Strategy Series: Transforming Developers into AI Architects with Google Cloud
This series emphasizes that data strategy and AI strategy are now intertwined, guiding developers on transforming into AI architects with Google Cloud. It provides best practices for building effective AI agents.
Announcing the MCP Toolbox Java SDK
Google Cloud announces the MCP Toolbox Java SDK, simplifying the integration of enterprise AI agents with mission-critical databases. This tool update reduces the need for custom glue code in agentic systems.
Azure
(14 articles)Even simpler to Safely Execute AI-generated Code with Azure Container Apps Dynamic Sessions
Azure Container Apps Dynamic Sessions make it simpler to safely execute AI-generated code by providing isolated environments. This is a critical security and development feature for AI agents.
Draft & Deploy on Azure Firewall
Azure Firewall Policy introduces a new Draft & Deploy feature, streamlining policy management and significantly reducing deployment time. This enhances security operations and efficiency.
Azure Databricks Lakebase
Azure Databricks Lakebase is now generally available, offering a managed PostgreSQL environment with next-gen storage/compute separation for OLTP. This provides instant availability, clones, and scale-to-zero for data workloads.
Azure Policy faster enforcement and retirement of login/logout workaround
Azure Policy now enforces Resource Manager mode policies within 5 minutes, significantly improving responsiveness and retiring previous workarounds. This enhances compliance and security posture management.
Migrating to the next generation of Virtual Nodes on Azure Container Instances (ACI)
This article discusses migrating to the next generation of Virtual Nodes on Azure Container Instances (ACI). It's important for users to understand these updates for performance and best practices.
Inference at Enterprise Scale: Architecting for Cost, Latency, and Efficiency
This article provides architectural guidance for enterprise-scale AI inference, focusing on optimizing for cost, latency, and efficiency. It's a critical best practice for deploying AI workloads in the cloud.
Instant access incremental snapshots: Restore without waiting
Azure introduces instant access support for incremental snapshots of Premium SSD v2 and Ultra Disk, enabling immediate creation, disk restore, and production-ready performance. This significantly improves data recovery and business continuity.
Powering secure private network streaming to Fabric with Eventstream connectors
Microsoft Fabric introduces Eventstream connectors for secure private network streaming, enabling real-time data access for business decisions. This enhances data integration and security for analytics platforms.
Update to required permissions for Semantic Models in Fabric Data Agents
Microsoft announces an update to required permissions for interacting with semantic models in Fabric data agents. This is important for maintaining security and access control for data professionals.
Unlocking document understanding with Mistral Document AI in Microsoft Foundry
Microsoft Foundry integrates Mistral Document AI to unlock document understanding, addressing challenges with unstructured data. This is a significant AI tool update for enterprises dealing with large volumes of documents.
Azure Developer CLI (azd): One command to swap Azure App Service slots
The Azure Developer CLI (azd) now includes a command to swap Azure App Service slots, making deployment slot swaps fast and intuitive. This improves DevOps workflows for Azure developers.
Azure Databricks update workspace network configuration
Azure Databricks workspaces now support updating network configurations, offering greater flexibility for Azure Databricks deployments. This is a key update for managing network settings in data analytics environments.
ExpressRoute Gateway Microsoft initiated migration
Microsoft is performing automated upgrades to ensure ExpressRoute gateways use the Standard IP SKU, enhancing gateway reliability. This is an important update for network architects using ExpressRoute.
Inference at Enterprise Scale: Why LLM Inference Is a Capital Allocation Problem
This article frames LLM inference as a capital allocation problem, highlighting the challenges of serving AI models efficiently. It's crucial for understanding the cost implications of large-scale AI deployments.
Cloudflare
(14 articles)From the endpoint to the prompt: a unified data security vision in Cloudflare One
Cloudflare One unifies data security from endpoint to prompt with new features like RDP clipboard controls, on-device DLP, and Microsoft 365 Copilot scanning. This provides a comprehensive security solution for modern cloud environments.
Always-on detections: eliminating the WAF “log versus block” trade-off
Cloudflare introduces Attack Signature Detection and Full-Transaction Detection for continuous, high-fidelity security insights without manual WAF tuning. This significantly improves web application security by identifying successful exploits proactively.
Mind the gap: new tools for continuous enforcement from boot to login
Cloudflare's mandatory authentication and independent MFA ensure continuous security enforcement from machine boot to sensitive resource access. This strengthens identity and access management in cloud environments.
Defeating the deepfake: stopping laptop farms and insider threats
Cloudflare One partners with Nametag to combat identity fraud and insider threats by requiring identity verification during onboarding and continuous authentication. This enhances cloud security by verifying user identities.
Moving from license plates to badges: the Gateway Authorization Proxy
Cloudflare's Gateway Authorization Proxy now supports identity-aware policies for clientless devices, securing virtual desktops and guest networks. This extends zero-trust security to a broader range of devices and use cases.
Stop reacting to breaches and start preventing them with User Risk Scoring
Cloudflare One incorporates dynamic User Risk Scores into Access policies, enabling automated, adaptive security responses. This allows for more nuanced and proactive security measures based on continuous behavior signals.
See risk, fix risk: introducing Remediation in Cloudflare CASB
Cloudflare CASB Remediation allows security teams to directly fix risky file sharing in Microsoft 365 and Google Workspace from Cloudflare One. This moves beyond visibility to active remediation of cloud SaaS security risks.
Introducing the 2026 Cloudflare Threat Report
Cloudflare's 2026 Threat Report highlights a fundamental shift towards industrialized cyber threats, including record DDoS attacks and sophisticated session token theft. This provides critical insights into the current threat landscape for cloud infrastructure.
Evolving Cloudflare’s Threat Intelligence Platform: actionable, scalable, and ETL-less
Cloudflare introduces new visualization, automation, and enrichment tools in its Threat Intelligence Platform to transform telemetry into instant security posture. This enhances threat hunting and security operations.
How Cloudy translates complex security into human action
Cloudy, Cloudflare One's LLM-powered explanation layer, is now part of Phishnet and API CASB, improving user decisions and SOC efficiency. This leverages AI for better security understanding and response.
From reactive to proactive: closing the phishing gap with LLMs
This article discusses how LLMs can help identify invisible weaknesses and close the phishing gap in email security. It highlights the proactive application of AI in cloud security.
Ending the "silent drop": how Dynamic Path MTU Discovery makes the Cloudflare One Client more resilient
The Cloudflare One Client now features Dynamic Path MTU Discovery, actively probing and adjusting packet sizes to eliminate problems from MTU differences. This improves network resilience and stability.
A QUICker SASE client: re-building Proxy Mode
Cloudflare transitioned its One Client to use QUIC streams for Proxy Mode, resulting in a 2x increase in throughput and significant latency reduction. This is a major performance improvement for SASE users.
How Automatic Return Routing solves IP overlap
Cloudflare's Automatic Return Routing (ARR) solves enterprise IP overlap challenges using stateful flow tracking. This simplifies network management and improves connectivity for complex cloud environments.
Cloud Native
(6 articles)The great migration: Why every AI platform is converging on Kubernetes
This article discusses why Kubernetes has become the converging platform for AI workloads, moving beyond stateless web services. It highlights a significant industry trend for cloud-native AI.
Declarative configuration is stable!
Key portions of the OpenTelemetry declarative configuration specification, including the JSON schema and YAML representation, have been marked stable. This is a significant milestone for OpenTelemetry users.
Support for Istio 1.27 ends on 30 March, 2026
Istio 1.27 support ends on March 30, 2026, urging users to upgrade to a newer version for continued security and bug fixes. This is a critical update for maintaining service mesh health.
KubeCon + CloudNativeCon Europe 2026 Co-located Event Deep Dive: KeycloakCon
This deep dive into KeycloakCon at KubeCon + CloudNativeCon Europe highlights the growing importance of identity and access management in cloud-native architectures. It's relevant for architects focusing on security and multi-cluster platforms.
KubeCon + CloudNativeCon Europe 2026 Co-located Event Deep Dive: Kubernetes on Edge Day
Kubernetes on Edge Day at KubeCon + CloudNativeCon Europe focuses on cloud-native technologies in distributed, resource-constrained edge environments. This is important for architects exploring edge computing strategies.
OTTL context inference comes to the Filter Processor
OpenTelemetry's Filter Processor now supports OTTL context inference, allowing users to write statements without worrying about internal telemetry contexts. This simplifies observability pipeline configuration.
Monitoring
(2 articles)Meet the new Bits AI SRE: Deeper reasoning, twice as fast
Datadog's Bits AI SRE receives upgrades including stronger reasoning for root cause analysis, expanded data sources, and new triage/remediation actions. This enhances AI-powered monitoring and incident response.
Use plain English to query your multi-cloud infrastructure in Resource Catalog
Datadog Resource Catalog now supports natural language querying, allowing users to search cloud resources and answer questions without complex syntax. This simplifies multi-cloud infrastructure management and observability.
DevOps
(6 articles)Why is your Kubernetes cluster adding nodes when the dashboards look fine?
This article explores common reasons why Kubernetes clusters might unexpectedly add nodes despite dashboards appearing normal, often due to bursty workloads like AI inference. It's crucial for Kubernetes performance and cost optimization.
NanoClaw can stuff each AI agent into its own Docker container to deal with OpenClaw’s security mess
This article discusses how NanoClaw can containerize AI agents using Docker to mitigate security issues found in OpenClaw. It highlights important security and isolation practices for AI agent deployments.
Is AI killing open-source software?
This article delves into the complex relationship between AI and open-source software development, exploring potential impacts and challenges. It's an important industry trend for developers and maintainers.
Open-source coding agents like OpenCode, Cline, and Aider are solving a huge headache for developers
This article highlights how open-source AI coding agents are addressing challenges faced by developers in managing LLM economics and proliferation. It's a significant trend in developer tooling.
Nearly half of all companies now use Rust in production, survey finds
The 2025 State of Rust Survey reveals that nearly half of all companies now use Rust in production. This indicates a growing industry trend for the programming language in enterprise environments.
High-performance envelope encryption at Ariso.ai with Vault
Ariso.ai adopted HashiCorp Vault's Transit engine for high-performance envelope encryption, securing its multi-tenant AI assistant. This showcases a real-world application of a popular DevOps security tool.
