Cloud · Top stories
AWS introduces simplified getting started experience for new builders
AWS has launched a new simplified onboarding experience for new users, particularly those working with AI, by offering sensible defaults, identity provider sign-up, and no credit card requirement to start. This change aims to reduce initial configuration hurdles and allow builders to deploy projects faster, with options to activate advanced features as needed.
SeaVerse Reduces Infrastructure Costs by 60% Using GKE Agent Sandbox
SeaVerse, a gaming startup focused on playable AI experiences, adopted Google Kubernetes Engine (GKE) and GKE Agent Sandbox to manage its AI workloads. This implementation reduced the company's infrastructure costs by up to 60% while providing strong isolation and low latency for dynamic, multi-tenant sandbox environments.
Dropbox Evolves Riviera Platform to Support AI Workloads and 300+ File Formats
Dropbox has expanded its Riviera content processing platform to handle over 300 file formats and more than 100 transformation capabilities, supporting AI workloads for products like Search, Replay, Sign, and Dash. This evolution allows AI features to reuse existing content processing infrastructure, enabling functions such as AI-powered summaries and content indexing.
AWS STS simplifies session token size limits to 4,096 bytes and adds monitoring
AWS Security Token Service (STS) has replaced its two session token size limits with a single 4,096-byte limit, providing more space for session policies and tags. STS now reports session token size in API responses, CloudWatch metrics, and CloudTrail events, allowing users to monitor and optimize token usage.
Amazon Cognito Introduces Multi-Region Replication for Enhanced Authentication Resilience
Amazon Cognito now offers Multi-Region Replication (MRR), automatically replicating user pools across AWS Regions with near-real-time synchronization. This feature simplifies building resilient authentication systems by handling user data synchronization, consistency, and failover, reducing operational overhead for businesses scaling globally.
Google Cloud Introduces Granular Session Management Tools with Automation and Group Targeting
Google Cloud has rolled out new session management tools that offer native, granular controls, moving beyond broad administrative settings. These updates include programmatic policy configuration via Terraform, gcloud, and APIs, along with the ability to target policies using Google Groups and apply controls to specific applications.
Amazon Bedrock Introduces Prompt Caching to Reduce Costs and Latency
Amazon Bedrock now offers prompt caching, which can reduce input token costs by up to 90% and lower time-to-first-token (TTFT) when repeatedly sending the same context to foundation models. This feature allows Bedrock to store and reuse partially processed input, avoiding redundant computation for subsequent requests with matching prefixes.
Cloudflare introduces new granular access controls and roles for Workers
Cloudflare has released new access controls for its Workers platform, allowing users to assign specific access levels to individual Workers. This update includes four new roles and the ability to apply permissions at the Developer Platform, product, or resource level, enhancing security and adherence to the principle of least privilege.
Google Cloud Dataflow Adds Pause/Resume for Batch Jobs and NVIDIA RTX PRO 6000 GPU Support
Google Cloud Dataflow introduced Pause/Resume functionality for batch jobs and added support for NVIDIA RTX PRO 6000 Blackwell GPUs. These updates allow users to recover from job failures and dynamically reallocate compute resources, while also providing enhanced processing power for AI inference workloads.
BigQuery Introduces New Augmented Analytics Table-Valued Functions
Google's BigQuery now includes a suite of augmented analytics Table-Valued Functions (TVFs) that automate complex data analysis. These functions integrate AI, ML, and statistical methods to diagnose metric changes, uncover trends, and isolate the impact of business decisions, operating directly within BigQuery.
Maryland Data Center Developers Offer $110M Community Benefits Package in Frederick County
Data center developers in Frederick County, Maryland, have proposed a $110 million community benefits package, including a new elementary school and water reclamation system, to gain approval for the Frederick Digital Campus. This initiative suggests a trend where tech companies are offering substantial local investments to address community concerns and secure approvals for large AI data centers.
Netflix Reworks Conductor Workflow Engine for Increased Scale and Efficiency
Netflix re-engineered its Conductor workflow orchestration engine to handle 420 million monthly workflow executions and support workflows 10 times larger. The redesign addresses bottlenecks in workflow evaluation and task coordination, improving performance and scalability for Netflix's internal operations.
TwelveLabs Marengo Embed 3.0 now available in Amazon Bedrock Knowledge Bases
TwelveLabs Marengo Embed 3.0 is now generally available as an embedding model within Amazon Bedrock Knowledge Bases. This integration allows for natural language search across video, audio, and image content, simplifying the process of building semantic search capabilities for multimodal assets.
Amazon EBS Volume Clones now support cross-account copying and re-encryption
Amazon EBS Volume Clones now allow users to copy EBS volumes across different AWS accounts and optionally re-encrypt them with an AWS KMS key in the target account. This feature enables developers to use production data for development and testing in isolated secondary environments.
Massachusetts Mandates Clean Power for New Data Centers Over 25 MW
Massachusetts Governor Maura Healey issued an executive order requiring new data centers larger than 25 megawatts to provide 100% clean power or contribute to a ratepayer protection fund. This move makes Massachusetts the third state in recent months to impose new restrictions on data center development, reflecting a broader trend of increased scrutiny on the industry.
AlloyDB Omni RPM Orchestrator for PostgreSQL is Generally Available
Google has announced the general availability of the AlloyDB Omni Red Hat RPM orchestrator, launching alongside AlloyDB Omni version 18.3.0. This release provides production-ready security, resiliency, and low-downtime operations for PostgreSQL workloads on virtual machines and bare-metal servers, integrating Google's AI capabilities.
AWS introduces Ray Serve Deep Learning Containers for TorchServe inference workloads
AWS has launched Ray Serve Deep Learning Containers (DLCs) to support model inference workloads, particularly for users migrating from the unmaintained TorchServe. These DLCs provide pre-built, optimized Docker images with a complete inference stack, including PyTorch and Ray Serve, maintained by AWS. This offering addresses the operational burden of managing dependencies and security for deep learning model serving.
Cloudflare Workers' Module Registry Rewritten for Node.js Compatibility
Cloudflare has rewritten the module registry in workerd, the core component of its Workers runtime, to improve Node.js compatibility and standards adherence. This update allows for better resolution, loading, and caching of modules, enabling deployment of larger Node.js applications to Cloudflare Workers.
Big Tech companies consider Patagonia for AI data centers due to climate and energy
Amazon, Google, and OpenAI are exploring Patagonia, Argentina, as a location for new AI data centers, drawn by its cool climate and abundant energy resources. The region's potential is hindered by a lack of high-speed networking infrastructure and political uncertainty. OpenAI has confirmed plans for a $25 billion, 500MW clean energy data center in the country.
Thailand requests pause on 49 data center builds for new regulatory framework
Thailand's government has asked data center operators to suspend 49 ongoing construction projects while it develops a comprehensive legal framework for the industry. This move addresses concerns over power and water usage and aims to create unified legislation for data centers, which currently operate under limited specific laws.
Cloudflare Dominates European CDN Market, Serving Nearly 90% of CDN-Using Companies
A study of European companies using Content Delivery Networks (CDNs) found that 89.6% of them use Cloudflare. This indicates Cloudflare's significant market share among European businesses that have adopted CDN services.
Broadcom removes public access to VMware VDDK downloads, hindering migrations
Broadcom has removed public download access for the VMware Virtual Disk Development Kit (VDDK), a library essential for many migration tools. This action creates an unexpected roadblock for customers attempting to migrate workloads from VMware vSphere to other platforms.
Netflix Adopts Open-Source Apache Flink Autoscaler for 30,000+ Streaming Jobs
Netflix is migrating over 30,000 streaming jobs to the open-source Apache Flink Autoscaler, moving away from its previous cluster-level autoscaling solution. This change addresses limitations in handling complex, stateful pipelines with varied operator processing requirements, leading to significant cost reductions.
Azure API Management Standard v2 Tier Gains Zone Redundancy
Microsoft has added zone redundancy to the Standard v2 tier of Azure API Management, a feature previously exclusive to Premium tiers. This change allows users to protect against datacenter-level failures at a lower cost, though it requires creating a new instance for existing deployments.
VMware to Release Updated vSphere Standard Amid SMB Pricing Concerns
VMware will release an updated version of vSphere Standard, addressing concerns from small-to-medium businesses (SMBs) regarding the high cost and bundled nature of its flagship VMware Cloud Foundation (VCF) offering. This move comes after Broadcom's acquisition led to subscription-based bundles that many SMBs found unaffordable, with sales teams reportedly pushing VCF over other options.
Tottenham Hotspur reduces virtualization licensing fees by 85% with HPE Morpheus VM Essentials
Tottenham Hotspur replaced VMware with HPE’s Morpheus VM Essentials virtualization software in its stadium, resulting in an 85% reduction in licensing fees. This migration is part of a broader data center and network infrastructure redesign, highlighting a trend of organizations seeking alternatives to VMware following Broadcom's acquisition.
Google Cloud Data Analytics, BI, and Database Teams Announce New Features
Google Cloud has introduced several updates across its data analytics, business intelligence, and database services, including stateful processing for BigQuery continuous queries, a synthetic data generator for Managed Service for Kafka, and faster Dataflow pipeline updates. These features aim to enhance real-time data processing, simplify testing, and improve pipeline management for users.
Equinix Partners with Nvidia and Together AI for New AI Inference Exchange Service
Equinix announced a partnership with Nvidia and Together AI to launch Equinix Inference Exchange, a service allowing customers to run AI models on Together AI's open-source cloud platform within Equinix data centers. This initiative positions Equinix to capitalize on the growing demand for AI infrastructure by providing colocation services for AI workloads, leveraging its global network of 281 facilities.
Reliance Jio Expands Cloud PC Service to All Indian Internet Users
Reliance Jio has made its JioPC cloud PC service available to all internet users in India, regardless of their provider. This expansion allows older computers to access AI-ready computing power via the cloud, potentially extending PC lifecycles and offering an alternative to hardware upgrades.
Proposed London Datacenter Project Faces Scrutiny Over High Carbon Emissions
A proposed "hyperscale" datacenter in outer London, the East Havering Data Centre Campus (EHDCC), is projected to generate 1.2 million tonnes of carbon dioxide annually once fully operational, equivalent to 27,000 flights from London to New York. The developer, Digital Reef, stated the project does not align with net-zero goals, raising concerns about its environmental impact and electricity consumption.
Microsoft Introduces Azure Multicloud Interconnect for Simplified AWS Connectivity
Microsoft has launched Azure Multicloud Interconnect, a new service designed to simplify private network connectivity between Azure and AWS environments. This collaboration between Microsoft and AWS uses the Open API specification to provide high-bandwidth, predictable, and secure connections, reducing the complexity of managing multicloud infrastructure.
AWS Graviton5-powered EC2 R9g and R9gd instances are now generally available
Amazon EC2 R9g and R9gd instances, powered by AWS Graviton5 processors, are now generally available, offering up to 25% better compute performance than previous Graviton4-based instances. These new memory-optimized instances provide faster memory, higher network and EBS bandwidth, and a larger L3 cache, making them suitable for memory-intensive workloads like databases and real-time analytics.
Microsoft Expands Foundry Model Router to 28 Regions, Updates Model Pool
Microsoft expanded its Foundry Model Router from two to 28 regions for global deployments and refreshed its model pool, adding Anthropic Claude Opus 4.8 and the GPT-5.6 family while removing older models. This update automatically applies to default configurations, potentially altering application behavior without redeployment, which could impact stability for some users.
Meta Tests Robots for Data Center Operations to Reduce Human Labor
Meta is testing robots from vendors like Watney Robotics, Kinova, and ABB to perform tasks such as plugging cables, resetting servers, and power cycling devices in its data centers. This initiative aims to reduce human labor costs as Meta's AI infrastructure expands, potentially replacing a significant portion of some workers' tasks.
Amazon ECS Express Mode simplifies container deployment on AWS Fargate
Amazon Web Services (AWS) has launched ECS Express Mode, a new interface for Amazon Elastic Container Service (ECS) that simplifies the deployment of containerized applications. This mode automates the setup of load balancers, networking, IAM roles, and scaling policies, allowing developers to deploy an HTTPS service on Fargate with just a container image and two IAM roles.
Cloudflare Workers Now Support Inbound TCP Connections and gRPC Protocol
Cloudflare Workers can now accept inbound TCP connections, expanding their capabilities beyond HTTP-only server functions. This update introduces a new `connect(socket)` handler and initially supports gRPC, enabling full-duplex communication for various TCP-based protocols.
AWS Management Console Private Access Now Supports VPCs Without Internet Connectivity
AWS Management Console Private Access is now generally available, allowing organizations to route all console traffic through AWS PrivateLink VPC endpoints without requiring internet connectivity. This update enables regulated industries to access the AWS Management Console from isolated network environments, addressing a previous limitation where static assets and console-only APIs still needed internet access.
EPA exempts islanded data center power generation from Clean Air Act Acid Rain Program
The U.S. Environmental Protection Agency (EPA) issued guidance clarifying that power generation facilities not connected to a public electricity grid, often used by data centers, are exempt from the Clean Air Act's Acid Rain Program. This change aims to provide greater flexibility for data center developers in building new facilities and supports the administration's agenda for artificial intelligence dominance and utility price stability.
Route 53 Files Launches to Enable DNS Management via Standard UNIX File System Operations
Route 53 Files is a new service that allows AWS Route 53 hosted zones to be accessed and managed as file systems using standard UNIX tools. This development simplifies DNS record management by integrating it with existing file system operations, making changes automatically reflected across Route 53 and connected compute resources.
Proton Mail and other Proton services experience widespread outage
Proton services, including Proton Mail, Proton VPN, and Lumo, experienced a major outage affecting multiple platforms. The company confirmed the issue and implemented a fix, with most services recovering for users. This outage impacts users relying on Proton's privacy-focused ecosystem for email, VPN, and other digital tools.