All news in category “Vendor and Hyperscaler Watch”

🔒 Amazon Bedrock expands its Reserved service tier to provide predictable tokens‑per‑minute capacity for mission‑critical workloads. The tier lets customers reserve prioritized compute and separately configure input and output tokens‑per‑minute to match asymmetric usage patterns and control costs. When reserved capacity is exceeded, traffic automatically overflows to the pay‑as‑you‑go Standard tier to avoid interruptions. Reserved access is available today for Anthropic Claude Opus 4.5 and Claude Haiku 4.5, with 1‑month or 3‑month reservations billed monthly per 1K tokens‑per‑minute.

🚀 Amazon Managed Workflows for Apache Airflow (MWAA) is now available in the AWS Region Asia Pacific (Thailand). The managed service delivers the familiar Apache Airflow orchestration platform with AWS-managed infrastructure, offering improved scalability, availability, and security while removing the operational burden of maintaining clusters. Customers in Thailand can deploy workflows closer to their data, helping reduce latency and address data residency needs. Consult the documentation and AWS region table to plan deployments and verify service limits.

🔒 IBM and Palo Alto Networks are partnering to deliver a unified, AI-powered cybersecurity foundation across Northern Europe, helping enterprises reduce tool sprawl, improve visibility and accelerate compliance. Their integrated stack—Cortex XSIAM, Cortex Cloud, Prisma Access and IBM consulting—secures cloud, AI pipelines and hybrid work while automating SOC workflows. The program targets measurable ROI, faster detection and simplified policy management aligned to NIS2, DORA and the EU AI Act.

🔁 AWS Outposts racks now support up to 10 isolated local gateway (LGW) routing domains per Outpost. Each domain has independent route tables, LGW VIF Groups, and BGP sessions, enabling traffic separation and preventing cross-domain routing while allowing both Customer-owned IP (CoIP) and Direct VPC Routing (DVR) on the same hardware. You can configure multiple LGW routing domains via the AWS Management Console or AWS CLI. This capability is available on second-generation Outposts racks at no additional charge.

🖧 AWS now supports multiple local gateway (LGW) routing domains on Outposts racks, allowing you to create up to 10 isolated routing domains per Outpost. Each domain has independent route tables, its own LGW VIF Group, and dedicated BGP sessions to on‑premises networks, enabling traffic separation and simultaneous use of Customer-owned IP (CoIP) and Direct VPC Routing (DVR). The capability is available on second‑generation Outposts racks at no additional charge and can be configured through the AWS Management Console or the AWS CLI.

🚀Palo Alto Networks partnered with Google Cloud to replace a brittle single-tenant data pipeline model with a unified, multi-tenant Unified Data Platform powered by Dataflow, Pub/Sub and BigQuery. The migration consolidated more than 30,000 pipelines into a shared, autoscaling platform that processes billions of events daily. The change delivered roughly 30% compute cost savings, faster onboarding, and reduced operational overhead, enabling engineers to refocus on analytics and threat detection.

🚀Second-generation AWS Outposts racks are now supported in the South America (São Paulo) and Europe (Stockholm) Regions. Outposts racks extend AWS infrastructure, services, APIs, and tools into on-premises data centers and colocation facilities to provide a consistent hybrid experience. Customers can order racks connected to these Regions to optimize for latency and data residency requirements while maintaining centralized application management.

🔍 Amazon S3 Storage Lens is now available in AWS GovCloud (US) Regions, providing organization-wide visibility into object storage usage and activity. S3 Storage Lens delivers cost, data protection, and performance metrics to identify inefficient access patterns, incomplete multipart uploads, and buckets with non-current object versions. Free basic metrics include 14 days of history while advanced metrics offer extended retention and greater detail.

🤖 This recap of the Agent Factory holiday special summarizes practical guidance on model fine-tuning, with a focus on reinforcement learning (RL) and Google’s TPU infrastructure. Hosts Shir Meir Lador and Don McCasland speak with Kyle Meggs from the TPU Training Team about when to fine-tune, the distinction between pre‑training, SFT, and RL, and why specialized workloads benefit from hosted solutions like MaxText on TPUs. The post also demonstrates a GRPO demo using Pathways, vLLM, and Tunix components to show RL at scale.

🔍 CrowdStrike outlines an architectural approach to enable agentic defense across the Falcon platform. The blog highlights Enterprise Graph for semantic data unification, Charlotte AI expert agents for native reasoning, and Charlotte Agentic SOAR for adaptive orchestration. It stresses governed, auditable execution and the ability to build custom agents with Charlotte AI AgentWorks. The aim is a real-time digital twin so agents and analysts share a single, continuously updated context to accelerate triage and response.

✉️ Google has begun rolling out an option that lets users change their primary @gmail.com address to a new @gmail.com address. When changed, the previous address becomes an alias and continues to receive mail; account data (photos, messages, and existing emails) is preserved and accessible. You can sign in with either the old or new address, revert to the prior address at any time, and Google limits creation of additional new addresses for the same account over a 12‑month period; the new address cannot be deleted.

📦 Amazon S3 on Outposts is now supported on second-generation AWS Outposts racks, offering 196 TB, 490 TB, and 786 TB storage tiers for on‑premises workloads. The update enables customers to use the same S3 APIs, security controls, and access management for local data residency, low-latency access, and on-site processing. It is available in all Regions and countries where second-generation racks are offered. Customers can select tiers optimized for production, backup, or archival use cases while maintaining familiar S3 management and features.

🚀 AWS Deadline Cloud integrates with Foundry Nuke CopyCat, enabling machine learning training jobs for visual effects to run directly on cloud render farms. Artists can submit CopyCat training jobs to scale workloads, run multiple trainings in parallel, and free local workstations for creative work. Training and render jobs are tracked together in the Deadline Cloud interface for unified project monitoring. The integration is available in all AWS Regions where Deadline Cloud is supported.

🚀 Microsoft released the Copilot Studio extension for Visual Studio Code, enabling developers to build and manage Copilot Studio agents directly within the editor. The extension lets teams pull full agent definitions locally, edit components with IDE features like syntax highlighting and IntelliSense-style completion, and preview or compare changes against the cloud. It supports Git versioning, CI/CD integration, and works with AI coding assistants to speed development; the extension is free on the VS Code Marketplace and has been downloaded over 13,000 times.

🚀 Amazon Aurora PostgreSQL, Amazon Aurora DSQL, and Amazon DynamoDB serverless databases are now accessible directly from v0 by Vercel, letting developers build full-stack applications and connect to AWS databases using natural language prompts. v0 provides an end-to-end setup experience to create or link AWS accounts, with new accounts receiving access to all three databases and $100 USD in credits. Serverless options scale to zero and are available in seven AWS Regions, reducing operational overhead for prototypes and production AI-driven applications.

🧩 AWS Clean Rooms now supports parameters in PySpark analysis templates, allowing template authors to define input values that collaborators supply at job submission time without editing the template code. When a collaborator is approved to run an analysis, they submit parameter values directly to the PySpark job, enabling reusable templates and faster iteration. This feature lets partners vary time windows, geographic regions, and other inputs dynamically to adapt analyses. It supports collaboration across companies on AWS or Snowflake and helps accelerate time-to-insights for use cases like advertising attribution.

🚀 BigQuery now supports managed third‑party generative AI inference (Preview) for open models from Hugging Face and Vertex AI Model Garden, enabling SQL-native deployment and inference. With a single CREATE MODEL statement you can provision and configure compute, control lifecycle with endpoint_idle_ttl and ALTER MODEL, and run inference via AI.GENERATE_TEXT or AI.GENERATE_EMBEDDING. BigQuery automates resource cleanup and integrates cost controls to reduce operational overhead.

🎉 Fortinet announced its 2026 Customer Excellence Award winners, recognizing organizations that have embedded security into core business strategy and delivered measurable outcomes. Awardees include Monolithic Power Systems, Wendy’s, Group 1 Automotive, Marvell Technologies, and Thames Water, showcasing advances in Secure SD‑WAN, OT protection, and security automation. The program highlights how the Fortinet Security Fabric enables consolidation, scalability, and operational resilience across industries.

🔒 Fortinet's Education Outreach program partners with Latinas in Cyber (LAIC) to increase representation of Latina women in cybersecurity through mentorship, practical training, and career pathways. Participants report that Fortinet's self-paced coursework and hands-on labs built technical confidence and clarified real-world security roles. Complimentary exam vouchers enabled candidates to pursue Fortinet certifications aligned with employer needs, helping translate training into tangible opportunities and career advancement.

🚀 AWS has announced general availability of Amazon EC2 X8i instances, a new family of memory-optimized instances built on custom Intel Xeon 6 processors exclusive to AWS. X8i offers up to 43% higher overall performance, 1.5x more memory capacity (up to 6 TB) and 3.4x greater memory bandwidth compared to prior X2i instances. Designed for SAP HANA, large databases, data analytics and EDA, X8i is SAP-certified and available in 14 sizes including two bare-metal options. Instances are initially available in US East (N. Virginia), US East (Ohio), US West (Oregon) and Europe (Frankfurt) and can be purchased via Savings Plans, On-Demand or Spot.