DBLiner in CerebroSQL Next-generation enterprise data orchestration: right inside your DBMS.
A lightweight, fully automated PL/pgSQL-based framework for managing data loading, ETL/ELT processes, and building analytical data marts in PostgreSQL and Greenplum. Forget about heavy external infrastructure—manage data flows right where the data lives.
Architectural advantages (Why DBLiner?)
Traditional external orchestrators (Apache Airflow, Prefect) require dedicated servers, complex Kubernetes clusters, and Python environment support, while also incurring massive network overhead. DBLiner changes the game
01
Zero Infrastructure
The entire engine is written in clean, optimized PL/pgSQL and deploys within your DBMS in a couple of minutes
02
No unnecessary load
DBLiner core queries are highly optimized for maximum speed. They perform instantaneous queue polling without placing a load on the Master node, even in distributed Greenplum clusters
03
Native transaction control
The framework ensures session isolation at the database level by dynamically generating execution contexts for each worker (RUNNER_<ID>)
04
Freeze protection
A built-in hard timeout mechanism at the process step level. No more hung transactions or broken dblinks locking the DBMS system catalog
Seamless CI/CD and GitOps for Data Engineers
Develop locally — deploy to production seamlessly.
One of the main challenges with in-database solutions is the difficulty of migrating between environments. DBLiner solves this through the "Configuration as Code" concept:
-
Creation and debugging: You design, link, and test ETL processes in a test environment (Dev/Stage) using the CerebroSQL visual editor.
-
One-click export: Once testing is complete, DBLiner automatically exports the full declarative script for process creation and configuration.
-
Version control and deployment: The script is seamlessly transferred to your corporate Git repository and deployed to the production environment via standard CI/CD pipelines. No manual metadata migration required
Cascaded data archiving
Security and Lifecycle (Enterprise Features)
Removing or modifying data marts in a production environment entails risks. DBLiner protects your data: during the cascading deletion or update of metadata objects, the framework automatically migrates all legacy schemas, parameters, and schedules to specialized archive tables (*_arch) under a unified `archive_id`. You can always roll back the changes
Comparison with alternatives
DBLiner strikes a balance between the heavy Airflow and declarative dbt for the PostgreSQL/Greenplum ecosystem:
Возможности фреймворков | DBLiner + CerebroSQL | Apache Airflow | dbt Core |
|---|---|---|---|
CI/CD / Git Concept | Yes (via script export) | Yes (Python files) | Yes (YAML configs) |
Control / Monitoring | CerebroSQL Visual UI (development) | Airflow web console | CLI only (in the free version) |
Custom scheduler | Yes (Built-in queue engine) | Yes | No (external Cron required) |
Where the code executes | Inside the DBMS (max. speed) | On external workers | Inside the DBMS |
Infrastructure footprint | Zero (DBMS only) | Heavy (Celery, Redis, DB) | Lightweight (CLI utility) |
Ready to get your databases in order?Start using DBLiner within the unified CerebroSQL platform—completely free of charge—right now.