DEV Community

Cover image for How to Read Parquet Files in SSIS using ZappySys ODBC (DuckDB)
Daniel ZS
Daniel ZS

Posted on

How to Read Parquet Files in SSIS using ZappySys ODBC (DuckDB)

Introduction

Reading Parquet files directly in SSIS can be challenging without the right connector. Using ZappySys ODBC PowerPack with the DuckDB (Parquet) connector, you can query Parquet files using SQL and integrate them seamlessly into SSIS data flows.

This guide shows the fastest way to set it up.


Steps

1. Create an ODBC Data Source

Open ODBC Data Sources (64-bit) and create a new System DSN so SSIS can access it.

2. Select ZappySys JDBC Bridge Driver

From the driver list, choose ZappySys JDBC Bridge Driver and continue.

3. Configure DuckDB (Parquet)

Set the connection to DuckDB in memory mode and point it to the DuckDB JDBC driver file.

Test the connection to confirm everything works.

4. Validate Parquet Access

Use the Preview tab to run a SQL query against a Parquet file and confirm data is returned.

5. Create SSIS Project

Open Visual Studio, create a new Integration Services Project, and add a Data Flow Task.

6. Add ODBC Source

Inside the Data Flow, add an ODBC Source and create a new ODBC Connection Manager using the DuckDB DSN.

7. Query Parquet Data

Choose SQL Command or Table name, preview the data, and confirm the schema.

8. Run the Package

Connect the ODBC Source to any destination and execute the package to process Parquet data in SSIS.


Conclusion

By combining SSIS with ZappySys ODBC PowerPack and DuckDB, you gain a powerful and flexible way to read Parquet files using SQL. This approach enables automated ETL workflows, analytics pipelines, and data exports without complex custom code.

👉 Full tutorial with screenshots and advanced examples

Top comments (0)