You define
- Source query SQL and target load SQL
- Full or incremental load
- Dependencies between jobs
- Start point for incremental queries
- Masking policies for sensitive columns
SOFTWARE
DBridge
DB Bridge
A batch engine that migrates and integrates database and API data through rules, and verifies that loaded data matches the source.
Migration and Integration Built from Rules
Migration and integration have usually meant a new program for every interface, with run order, retries, and result checks managed by hand. DBridge registers each interface as a batch rule, and the engine handles loading, retries, and reconciliation.
The engine performs
- Change window — query range calculated on every run
- Execution order — sequential and parallel runs based on dependencies
- Row-count check — the run fails when the difference between read and loaded counts exceeds the tolerance
- Failed-window retry — a failed window is read again on the next run
- Duplicate prevention — rules with a merge key leave no duplicates when the same window is re-run
DB→DB and API→DB Loads
Data from relational databases and HTTP APIs is loaded into relational databases. In the extended configuration, database data is also loaded into search indexes, vector databases, and graph databases under the same rule system.
HTTP APIs are used only as sources. DB→API delivery and real-time change data capture (CDC) are not supported.
Extraction to Verification in One Engine
Source extraction, target loading, reconciliation, and scheduled operation run in a single engine.
Extract source queries
- Source queriesQueries only on the source database (except pre- and post-processing SQL set in a rule)
- Incremental queriesChanges since the previous run
- API integrationDeclarative connection to external APIs that return JSON or XML and use key or header authentication
- Type conversionCross-DBMS type conversion with precision-loss warnings
Load target loading
- Full and incremental loadsLoad method set per rule
- Bulk loadingBulk load method for PostgreSQL, MySQL, and MariaDB targets
- Parallel loadingLarge jobs split and processed in parallel
- Search, vector, and graph loadsDatabase sources loaded under the same rule system in the extended configuration
- Data maskingSensitive columns masked by policy before loading
Verify consistency checks
- Row-count checkRows read compared with rows loaded
- Source reconciliationTarget re-queried after loading and compared with the source
- Mismatch tracingStaged reconciliation that locates mismatches
- Masking checkDetection of original values left in masked columns
- Verification ledgerTamper-evident record of reconciliation results
Operate batch operation
- Built-in schedulerScheduled and manual runs, pause and resume
- Dependency controlDownstream jobs run according to upstream results
- Duplicate-run protectionConcurrent runs of the same job blocked
- Run and change historyRun results and rule changes kept as history
Results that fail reconciliation are also recorded in the verification ledger.
Use Cases
One-off conversions and recurring integration run under the same rule system.
- Data conversionBulk conversion for system rebuild and migration projects
- Analytics warehouse collectionPeriodic collection from operational systems into an analytics warehouse
- External API ingestionExternal organization and service API data loaded into business databases
- Search index loadingBusiness data reflected in full-text search (extended configuration)
- Relationship graph loadingGraph data built for relationship analysis (extended configuration)
- Vector embedding loadingVector data built for semantic search (extended configuration)
Management Console
Explore batch flows, operations, database queries, and reconciliation in a synthetic ERP demo.
ERP batch flow · Dependencies 
The ERP screens are an actual DBridge demo using synthetic data. The API mapping is a separate synthetic example.
Screen Details
Supported Sources and Targets
- PostgreSQLSource · Target
- OracleSource · Target
- MySQL · MariaDBSource · Target
- MSSQLSource · Target
- HTTP APISource only
- Search index (FTS)Target only · extended configuration
- Vector databaseTarget only · extended configuration
- Graph database (GDB)Target only · extended configuration
Closed-Network Installation and Operation
An offline bundle installs without an on-site build, and neither installation nor operation requires an internet connection.
- Offline bundleInstallation components delivered with integrity checksums
- InstallationIntegrity check, then installation without an on-site build
- Encrypted secretsConnection passwords and API credentials stored encrypted with our crypto library (STRCrypt)
- Change historyWho changed a setting, when, and its before and after values
Deployment Environment
- Version
- 1.0
- Runtime environment
- Linux · Docker Compose
- Included components
- Batch engine · Web console · Metadata database
- Installation
- Closed-network offline bundle (no internet connection required)
Adoption Process
- Requirements analysis — integration direction, source and target systems, data volume, and run frequency
- Rule registration — connections, batch rules, and job dependencies
- Verification and operation — test runs to confirm results, then scheduled operation
Adoption Inquiry
We provide configuration and rollout guidance based on your requirements and operating environment. Demo and proposal requests are handled through the same contact.