Cloud · Top stories
AWS European Sovereign Cloud to Operate Without Global Backbone for Several Hours
AWS will temporarily disconnect its European Sovereign Cloud from the AWS Global Network backbone for several hours on October 24. This exercise aims to demonstrate the region's operational independence and align with EU cloud sovereignty frameworks, though it does not address legal sovereignty concerns.
Amazon Acquires DuckLabs, Integrates DuckDB into Amazon Aurora PostgreSQL
Amazon has acquired DuckLabs, the company behind the open-source analytical database DuckDB. DuckDB's creators will continue to lead the project within AWS, and the database will remain open-source under the MIT license, with the DuckDB Foundation maintaining supervision.
Vercel Introduces Daemons for Always-On Background Workers and General Availability of Python Support
Vercel launched Daemons, a new service for continuously running background workers, and made its Python support generally available. These additions expand Vercel's backend capabilities, allowing developers to run more complex, continuously processing workloads directly on the platform.
Google Cloud Enhances VPC Service Controls with Folder-Based Perimeters
Google Cloud has updated VPC Service Controls (VPC-SC) to allow folders to be added to perimeter membership, enabling projects within those folders to inherit security policies. This change simplifies the creation, maintenance, and scaling of network perimeters, particularly for managing autonomous agents and protecting data by default.
GitHub reports on availability improvements and Azure migration progress for August-September
GitHub released an availability report detailing progress on infrastructure improvements and migration to Azure during August and September. The company reduced shared-database pressure, added overload controls, and expanded failure containment for pull requests, while also moving more production workloads to Azure. These efforts aim to improve GitHub's overall availability and resilience.
Netflix implements tiered storage for TimeSeries data, moving older data to S3
Netflix has implemented two solutions for tiered storage of its TimeSeries data, moving older, less frequently accessed data from hot storage to Amazon S3. This change reduces storage costs while maintaining data availability for audits and investigations.
Netflix Rebuilds Ads Event Pipeline with Apache Flink for Live Streaming
Netflix re-engineered its Ads Event Publisher to support live streaming, de-duplication, and richer data signals. The update replaced a database lookup with real-time streaming joins in Apache Flink, enabling scalability for live content and removing a critical path dependency.
Netflix Automates Container Right-Sizing for Serverless Media Processing Platform
Netflix developed the Stratum Resource Tuner (SRT) to automatically right-size containers on its internal serverless media processing platform, Stratum. This system addresses the challenge of manual resource over-provisioning and configuration drift, which leads to increased operational costs and cognitive load for engineers.
Netflix details GPU infrastructure reliability for ML training and inference
Netflix developed methods to detect failing GPUs and network fabric, automatically remediate machines, and allow long-running jobs to survive hardware loss. This system ensures continuous operation for machine learning models that power personalization and generative AI initiatives.
Amazon S3 Now Available on Vercel Marketplace and v0 by Vercel
Amazon S3, including S3 Tables and S3 Vectors, is now natively available on the Vercel Marketplace and v0 by Vercel. This integration allows developers to provision and configure S3 storage directly within the Vercel platform, simplifying the process for web and AI applications.
Topgolf Migrates 12,000 VMs from VMware to VergeIO, Saving Over $11 Million
Topgolf moved 12,000 virtual machines from VMware to VergeIO's VergeOS, saving over $11.3 million in infrastructure costs. This migration was prompted by Broadcom's 400% price increase for VMware licenses, leading Topgolf to seek alternative solutions for its on-premises workloads.
Cloudflare D1 Scales to Millions of Databases, Supporting 65x Growth in Two Years
Cloudflare's D1 serverless database, based on SQLite, has scaled its control plane to support millions of databases, experiencing a 65x growth in live databases over two years. This scaling allows D1 to accommodate applications that use many small databases per user or tenant, addressing a shift in database architecture driven by AI-assisted development platforms.
Workday Opens HR and Finance Data to External AI Agents with Headless Platform
Workday has made its Trusted Agent Platform "headless," allowing developers to build AI agents that interact directly with HR and finance data using tools like Claude Code and Cursor. This change provides direct SQL access to live Workday data and ensures all agents operate under existing Workday permissions and audit trails, addressing data access control concerns.
AWS Enhances EventBridge Custom Event Buses with Cross-Account Sharing and New Features
AWS introduced enhanced custom event buses for Amazon EventBridge, enabling centralized sharing across accounts with features like ordered delivery, event replay, and content-based deduplication. This update addresses operational complexity and routing costs in multi-account environments, offering a new architecture for event-driven systems.
Dutch Tax Authority Halts Microsoft 365 Cloud Migration, Shifts to On-Premises and Open Source
The Netherlands' Tax and Customs Administration canceled its planned Microsoft 365 public-cloud rollout, opting instead for an on-premises and open-source approach. This decision aligns with a broader European focus on digital sovereignty, as indicated by a Linux Foundation report stating 94% of EU organizations prioritize it.
Meta Introduces New Open Networking Infrastructure for AI at OCP Summit 2026
Meta unveiled its next generation of open networking infrastructure at the 2026 Open Compute Project Global Summit, including liquid-cooled racks, a 102.4T switch platform, and a new multiplane fabric. These innovations support Meta's AI workloads and aim to enable more efficient, scalable, and sustainable AI infrastructure for the industry.
Yandex Cloud Experiences Outage at Third Data Center
Yandex Cloud reported an incident affecting its third data center, leading to service disruptions. This marks the third such incident for Yandex Cloud, indicating potential underlying infrastructure issues.
Battery storage now cheaper than open-cycle gas turbines for data centers
A Wood Mackenzie report indicates that four-hour duration battery storage is now less expensive than open-cycle gas turbines across 43 global markets. This shift impacts data center developers and utilities, as rising energy demand from AI data centers has increased gas turbine prices and procurement times.
East Fishkill Imposes Moratorium on Large Datacenters Amidst Community Backlash
East Fishkill, New York, has enacted a three-year moratorium on new datacenters larger than 20 megawatts following community opposition to a proposed 1-gigawatt AI datacenter. The temporary ban, effective until July 2029, allows officials to study the environmental impact of such facilities, highlighting growing local resistance to large-scale tech infrastructure.
Cloudflare Workers and Durable Objects now support on-demand CPU and memory profiling
Cloudflare has introduced on-demand CPU and memory profiling with flamegraphs for Workers and Durable Objects. This feature allows developers to identify exact functions consuming CPU or allocating memory, improving resource optimization in production environments.
Amazon ECS introduces automatic repair for failing GPUs and instances
Amazon ECS now automatically repairs failing GPUs and instances to improve application resilience. This update shifts some infrastructure failure recovery responsibility from users to ECS, simplifying operational management for applications running at scale.
Amazon SageMaker HyperPod integrates new Ray capabilities for distributed ML workloads
Amazon SageMaker HyperPod now includes new capabilities for Ray, an open-source framework for scaling distributed Python workloads. This integration simplifies the management of Ray clusters on HyperPod's infrastructure, providing built-in fault tolerance and improved observability for large-scale machine learning.
Google Cloud Launches ULL Solution with U4 Machine Family for Ultra-Low Latency Trading
Google Cloud has made its Ultra Low Latency (ULL) Solution generally available, introducing the new U4 machine family. This solution provides cloud-based infrastructure for high-frequency trading workflows, aiming to offer predictable performance and low latency for capital markets participants.
Grab Migrates Counter Service to Aerospike, Achieves 50% Lower P99 Latency
Grab migrated its Counter Service storage backend from a wide-column database to Aerospike, resulting in a 50% reduction in P99 read latency and a 66% decrease in on-disk data. This change impacts Grab's anti-fraud platform, which handles approximately one billion requests daily, by improving performance and reducing infrastructure costs.
Google signs 20-year agreement with Constellation for nuclear power plant upgrades
Google signed a 20-year power purchase agreement with Constellation to fund efficiency upgrades at six nuclear power plant sites. This deal aims to generate 890 megawatts of additional nuclear capacity to power Google's data centers, addressing increasing electricity demands from AI operations.
Google Cloud introduces serverless Lakehouse runtime catalog powered by Spanner
Google Cloud launched a serverless Lakehouse runtime catalog, built on Spanner, to address the need for a highly scalable and available managed catalog in lakehouse architectures. This new catalog adheres to the Apache Iceberg REST catalog specification, providing atomic commits and high availability for large-scale data operations.
Xeal Plans to Deploy 100,000 Nvidia GPUs at EV Charging Sites for Edge Inference
EV charging company Xeal will deploy over 100,000 Nvidia Hopper or Blackwell Ultra GPUs in "Latient Pods" across its 1,600+ US charging locations. This initiative aims to create an edge inference compute network by utilizing existing, underutilized electrical infrastructure at EV charging sites.
Survey: 90% of VMware Users Exploring Alternatives Due to Licensing Costs
A survey by Rimini Street found that 90% of VMware users are exploring alternatives, primarily driven by increased licensing costs following Broadcom's acquisition. The survey indicates that cost savings are a top priority for organizations re-evaluating their virtualization strategies.
Data Center Construction Spending Reaches Record $85 Billion Annual Pace
Private data center construction spending increased by 73% year-over-year to an annual pace of $84.95 billion, according to August Census Bureau data. This record growth signifies that data center buildouts now constitute 63% of all private office construction, highlighting their increasing importance in the overall construction sector.
Drax Datacentre Could Emit Double Gatwick's Annual Flight Emissions
An analysis by the Natural Resources Defense Council (NRDC) indicates that Drax's proposed 1.2GW datacentre could generate 8.6 million tonnes of CO2 annually by burning 4.9 million tonnes of wood. This projected emission level is nearly double the yearly CO2 output from all flights at Gatwick Airport, raising concerns about the environmental impact of AI datacentres powered by biomass.
Palo Alto Networks launches Cortex XCOR for AI-driven observability and automated incident response
Palo Alto Networks introduced Cortex XCOR, an AI-driven observability platform designed to automatically identify root causes and recommend fixes for system outages. This platform aims to shift incident response from manual dashboard monitoring to autonomous AI agent investigations, building on the acquisition of Chronosphere.
Tencent Secures 100,000 Offshore AI Chip Deal with Oracle for $7 Billion
Tencent reportedly signed a five-year, $7 billion deal with Oracle for 100,000 advanced AI chips in Southeast Asian data centers. This agreement allows Tencent access to high-end AI compute capacity unavailable in China, despite rising compute rental prices.
Google's Finnish data center project investigated for clearing forest without environmental assessment
Google's local operator in Finland, Tuike Finland, is under investigation for allegedly clearing over 300 hectares of forest for a new data center without a mandatory environmental impact assessment. This incident has prompted calls for the suspension of construction activities until legal compliance is confirmed.
Improper Redaction Reveals Google Data Center Water and Electricity Usage in Nebraska
Improper redaction in annual reports submitted to Nebraska's Department of Water, Energy, and Environment revealed specific water and electricity usage for Google's data centers. The reports, which Google had claimed as trade secrets, also disclosed expected tax refunds for 2025. This information provides insight into the environmental impact and financial incentives associated with large data center operations.
Proposed 725-hectare Anthropic AI datacenter in rural Queensland faces local opposition
A Singapore-based developer, Zerra DC, plans to build a 725-hectare datacenter in rural Queensland for Anthropic's Claude AI, requiring 2.16GW of power. Local residents and a petition with over 21,500 signatures oppose the project due to concerns about energy demand, noise, and the potential for boom-bust economic cycles similar to past resource projects.
Microsoft implements "biomimicry" program for data center landscaping at 20+ sites
Microsoft is integrating wetlands and native gardens at over 20 data center sites in the US and Germany to blend facilities into local ecosystems. This initiative, part of all new US projects, uses an Azure-based tool to assess environmental features and customize landscaping solutions. The program aims to address community concerns about environmental impact, despite some criticism regarding the company's energy deals.
FTL: A New Operating System for Clouds
FTL is a new operating system designed for cloud environments, allowing users to build custom OS instances as libraries. It aims to provide container isolation comparable to hypervisors while maintaining Linux binary compatibility and performance.
Public backlash against data centers spreads to Europe and Asia, impacting investments
Public opposition to data centers, initially prominent in the U.S., is now growing in Europe and Asia, leading to project delays and cancellations. This pushback is driven by concerns over power consumption, water usage, and land requirements, posing risks for data center investors globally.
SQL Server on Azure Local is now generally available for connected and disconnected operations
SQL Server on Azure Local is now generally available, allowing organizations to run SQL Server in environments with or without cloud connectivity. This release enables modernization of data infrastructure while maintaining control over data placement and supports bringing AI closer to data with Foundry Local.
Cloudflare launches Streamline developer playground for custom video pipelines
Cloudflare introduced Streamline, a new developer playground demonstrating how to build custom video pipelines using Cloudflare Stream, Workers, and Containers. Streamline allows developers to create dynamic video experiences like live annotations or burned-in subtitles by modifying video and publishing the output as a livestream or hosted video. This provides a framework for developers to implement bespoke video processing workflows on Cloudflare's platform.