AWS Data Pipeline
Legacy ETL orchestration
Closed to new customers since July 2024; use Glue, Step Functions, or MWAA instead. Data Pipeline defines scheduled data movement and processing flows between S3, RDS, DynamoDB, EMR… and runs on EC2/EMR resources it creates.
- Cost$ Low
- Difficulty●●●●●
- Costs money when idle?A little
- Managed by AWS
⚠️ Retired or retiring service
Open Data Pipeline on the map →
Closed to new customers since July 2024; use Glue / Step Functions / MWAA
Think of it asLike an old factory conveyor belt: it still runs for workshops already using it, but new workshops install more modern production lines.
When to use it
✓ Use it when
Only when maintaining legacy pipelines while waiting to migrate.
✗ Avoid it when
Any new data flow → Glue / Step Functions / Managed Workflows for Apache Airflow.
Common uses
- Maintain legacy DynamoDB-to-S3 backup jobs
- Plan the migration to Glue or MWAA
What does it cost?
- Pricing model: Charged by the number of activities/preconditions scheduled per month, plus the EC2/EMR resources the pipeline launches.
⚠️ Cost traps
EC2/EMR resources created by the pipeline keep incurring charges if they don't shut down on their own.
