All news with #apache airflow tag

Amazon MWAA Serverless for Apache Airflow Workflows

🚀 Amazon launched MWAA Serverless, a managed, serverless deployment option for Apache Airflow that eliminates infrastructure management and bills only for actual task compute time. Workflows can be authored as YAML configurations or Python DAGs and leverage over 80 AWS Operators from Airflow v3.0. Each workflow runs in isolation with distinct IAM permissions while the service automatically provisions and scales resources across supported regions.

Amazon MWAA Announces Support for Apache Airflow 3.0

🚀 Amazon Managed Workflows for Apache Airflow (MWAA) now supports Apache Airflow 3.0, enabling easier authoring, scheduling, and monitoring of complex workflows. The release introduces a redesigned UI and an event-driven scheduler that can trigger workflows directly from external events. The new Task SDK and Task Execution API reduce boilerplate, improve isolation, and limit direct metadatabase access. MWAA also adds Python 3.12 support and security fixes to enhance reliability.

Amazon MWAA Enables In-place Downgrades of Airflow

🔄 You can now perform in-place minor Apache Airflow version downgrades on Amazon Managed Workflows for Apache Airflow (MWAA). This feature lets administrators revert an MWAA environment to any other supported minor Airflow version without needing to recreate the environment. You can also launch new Airflow environments in all currently supported MWAA regions via the AWS Management Console with a few clicks. For details, see the Amazon MWAA documentation.