Transfer DuckDB to SQL Server
Fast, streaming database-to-database transfer with parallel processing
Terminal
.\FastTransfer.exe `
--sourceconnectiontype "duckdbstream" `
--sourceserver "your-source-server" `
--sourceuser "your-username" `
--sourcepassword "your-password" `
--sourcedatabase "your-source-db" `
--query "SELECT * FROM your_schema.your_table" `
--targetconnectiontype "msbulk" `
--targetserver "your-target-server" `
--targetuser "your-username" `
--targetpassword "your-password" `
--targetdatabase "your-target-db" `
--targetschema "your_table" `
--targettable "your_table" `
--parallelmethod "Ntile" `
--distributekeycolumn "id" `
--degree 10 `
--loadmode "Truncate"Source - DuckDB
DuckDB is a fast and efficient in-process analytical database. FastTransfer uses DuckDBStream to read multiple file formats with exceptional performance.
Features:
- •Direct reading of multiple formats (CSV, Parquet, JSON, XLSX)
- •Wildcard support to process multiple files
- •No installation required - embedded engine
Parallel Method - Ntile
Uses the distributed column and NTILE to build evenly sized chunks. Each thread exports a range from the column values; uniqueness is not required.
Available parallel methods with DuckDB:
Destination - SQL Server
FastTransfer uses SqlBulkCopy, Microsoft's native API for bulk loading. This method offers the best possible performance for SQL Server.
Loading method:
SqlBulkCopy API
Advantages:
- •Native SqlBulkCopy
- •High-performance bulk loading
- •Support for SQL Server-specific types
