Moving schedules and pipelines into self-hosted Airflow from cron, retired Hadoop schedulers, enterprise job schedulers and managed Airflow services, with coexistence and cutover planned.
Replacing crontabs spread across servers with Airflow DAGs: inventorying jobs, translating schedules to Airflow 3 timetables, adding retries, dependencies and alerting, handling scripts that need local files, and retiring cron host by host.
Leaving Oozie after its retirement to the Apache Attic: mapping workflows, coordinators and bundles to DAGs, timetables and assets, replacing Hadoop actions with Spark, Hive and SSH operators, and decommissioning the Oozie server.
Moving Luigi task pipelines to Airflow: Luigi targets and requires versus Airflow dependencies and assets, scheduling that Luigi leaves to cron, central history and retries, and porting tasks incrementally.
Moving data pipelines off BMC Control-M while keeping enterprise jobs running: the Control-M for Apache Airflow plug-in for a coexistence phase, translating calendars and dependencies, SLA monitoring, and the order to move job folders.
Taking DAGs from Amazon Managed Workflows for Apache Airflow to an Airflow you run: matching versions and constraints, scheduler and worker limits you outgrow, multi-team needs, AWS connections and secrets, and a parallel run before cutover.
Moving DAGs from Cloud Composer, now documented as Managed Service for Apache Airflow, to self-hosted Airflow: Composer generations and image versions, Google connections and Secret Manager, GCS DAG folders, and cutting over without double runs.
Running DAGs built on Astro Runtime on community Airflow: mapping Runtime versions to Airflow releases, replacing Astronomer-specific packages, rebuilding images on the official base, and what the platform team must now run itself.
When data workflows built as Step Functions state machines should move to Airflow: mapping states to tasks, keeping Lambda and Glue steps, running Step Functions from Airflow during the transition, and which workflows should stay event-driven in AWS.
Moving scheduling and dependencies out of Azure Data Factory into Airflow while keeping ADF copy activities where they work: running pipelines from Airflow, parameters, deferrable waits, and retiring ADF triggers.
Tell us what you're working on, whether it's a deployment, an audit, a security test or a cyber range. You'll speak with an engineer who can help you scope it.