AWS Marketplace pricing is a lot easier to understand now

Amazon Web Services just added AI Insights to Marketplace listings, so buyers can see pricing explained in plain language before they commit.

The update breaks down pricing units, scaling behavior, and multi-dimension pricing directly on the product page. That makes it easier to estimate what a product may cost without having to piece together the details yourself.

Amazon Aurora Serverless scales faster for bursty workloads

Amazon Web Services updated Aurora Serverless scaling so it can reach up to 12 ACUs within a second during scale-ups.

AWS also says it can scale up to 256 ACUs overall. The update is aimed at bursty, agentic AI, and unpredictable workloads.

AWS Identity Center is getting simpler for new org setups

Amazon Web Services also made IAM Identity Center account management optional for new organization instances, which trims down the setup surface for workforce access.

In the same update, AWS added support for replicating the Identity Center directory from a primary Region to additional Regions. That gives organizations a better way to keep workforce access available across Regions.

AWS Lambda now gives bigger network bandwidth without extra charges

Amazon Web Services updated AWS Lambda network bandwidth scaling, so functions outside a VPC with at least 2 GB of memory can scale bandwidth proportionally.

At 10 GB of memory, Lambda can now reach up to 3,000 Mbps. AWS says there’s no additional bandwidth charge for this.

AWS Glue Data Quality now removes pricing for anomaly detection

Amazon Web Services updated AWS Glue Data Quality with a new observation mode and removed pricing for anomaly detection in Glue ETL jobs.

The observation mode is designed to reduce false anomalies. At the same time, the pricing change makes continuous anomaly monitoring available without that extra cost barrier.

AWS Config now covers more resource types

Amazon Web Services expanded AWS Config coverage with 15 additional resource types across Amazon Bedrock, OpenSearch Serverless, SageMaker, and more.

The update improves visibility for auditing, inventory, and cost governance across all Regions where those resources are available. That gives teams a broader picture of what’s deployed.

Amazon CloudWatch now handles Prometheus collectors for you

Amazon Web Services announced managed Prometheus collectors in CloudWatch, which provision, scale, and collect metrics automatically.

The collectors support OpenTelemetry and Prometheus metrics for EKS, EC2, ECS, MSK, and OpenSearch Service workloads. That reduces the need to run and manage collector infrastructure yourself.

Amazon ECR can now store much larger image layers

Amazon Web Services increased the maximum Docker image layer size in Amazon ECR to 200 GB.

The update is meant to make packaging large assets like LLMs and datasets easier. It can also reduce the need to split layers or rely on external storage.

Amazon Redshift Data API is getting easier to use and cheaper to run

Amazon Web Services updated the Redshift Data API with long polling, ListSessions, and flexible batch execution modes.

Long polling reduces the number of polling calls, while ListSessions gives better visibility into active sessions. The flexible batch execution mode lets statements run independently with AUTO_COMMIT.

That helps improve ETL efficiency and lowers API chatter. For users, that means less overhead and better control over batch workloads.