Skip to main content
Version: 0.14

Advanced Scenarios

Advanced use cases including DuckDB integration, ClickHouse, SAP HANA, Teradata, and other complex transfer scenarios.

DuckDB Integration​

DuckDB provides powerful data transformation capabilities with FastTransfer.

DuckDB Memory - Parquet with Nested Fields to MSSQL​

Process complex nested JSON/Parquet files with DuckDB's in-memory engine:

.\FastTransfer.exe `
  --sourceconnectiontype "duckdb" `
  --sourceserver ":memory:" `
  --query "select * EXCLUDE(Risques, Interlocuteurs),unnest(interlocuteurs,recursive := true ) from (select day, unnest(contexte_chantier, recursive := true), unnest(detail_chantier, recursive := true) from read_parquet('D:\CLIENTS\ENEDIS\BI4ALL\CHANTIER\**\*.parquet', hive_partitioning=true)) t_row_unest" `
  --targetconnectiontype "msbulk" `
  --targetserver "localhost" `
  --targettrusted `
  --targetdatabase "ENDSBI4ALL" `
  --targetschema "bi4all" `
  --targettable "DuckChantiers_hpart" `
  --parallelmethod "DataDriven" `
  --distributeKeyColumn "day" `
  --loadmode "Truncate" `
  --degree 8
DuckDB Memory Mode

Use :memory: as the server to process data without persisting a DuckDB database file. Perfect for ETL transformations.

DuckDB Stream - Multiple CSV Files with DataDriven​

Process multiple CSV files in parallel using filename as distribution key:

.\FastTransfer.exe `
  --sourceconnectiontype "duckdbstream" `
  --sourceserver ":memory:" `
  --query "SELECT * exclude filename FROM read_csv('D:\temp\FastBCP_lineitem_Extract\lineitem*.CSV', filename=true)" `
  --targetconnectiontype "msbulk" `
  --targetserver "localhost" `
  --targettrusted `
  --targetdatabase "tpch_test" `
  --targetschema "dbo" `
  --targettable "lineitem" `
  --parallelmethod "DataDriven" `
  --distributeKeyColumn "filename" `
  --datadrivenquery "select file from glob('D:\temp\FastBCP_lineitem_Extract\*.CSV')" `
  --loadmode "Truncate" `
  --degree 16

DuckDB Stream - Single Large CSV with Random Method​

Process a single large CSV file with parallel Random distribution:

.\FastTransfer.exe `
  --sourceconnectiontype "duckdbstream" `
  --sourceserver ":memory:" `
  --query "SELECT * FROM read_csv('D:\temp\lineitem*.CSV')" `
  --targetconnectiontype "msbulk" `
  --targetserver "localhost" `
  --targettrusted `
  --targetdatabase "tpch_test" `
  --targetschema "dbo" `
  --targettable "lineitem" `
  --parallelmethod "Random" `
  --distributeKeyColumn "l_orderkey" `
  --loadmode "Truncate" `
  --degree 16

DuckDB Stream - CSV with Column Slicing​

Use calculated column slices for parallel distribution:

.\FastTransfer.exe `
  --sourceconnectiontype "duckdbstream" `
  --sourceserver ":memory:" `
  --query "SELECT * FROM read_csv('D:\temp\lineitem*.CSV') rcsv" `
  --targetconnectiontype "msbulk" `
  --targetserver "localhost" `
  --targettrusted `
  --targetdatabase "tpch_test" `
  --targetschema "dbo" `
  --targettable "lineitem" `
  --parallelmethod "DataDriven" `
  --distributeKeyColumn "(l_orderkey/5000000)::int" `
  --datadrivenquery "select distinct (l_orderkey/5000000)::int l_orderkey_group from read_csv('D:\temp\lineitem*.CSV') rcsv" `
  --loadmode "Truncate" `
  --degree 13

DuckDB Stream - CSV with Custom Format​

Specify CSV format parameters when auto-detection isn't sufficient:

.\FastTransfer.exe `
  --sourceconnectiontype "duckdbstream" `
  --sourceserver ":memory:" `
  --query "SELECT * FROM read_csv('D:\OpenData\wikidata\vector_database_wikipedia_articles_embedded.csv', auto_detect=false, delim=',', quote='""', escape='""', new_line='\n', skip=0, comment='', header=true, columns={'id': 'BIGINT', 'url': 'VARCHAR', 'title': 'VARCHAR', 'text': 'VARCHAR', 'title_vector': 'VARCHAR', 'content_vector': 'VARCHAR', 'vector_id': 'BIGINT'}) rcsv" `
  --targetconnectiontype "pgcopy" `
  --targetserver "localhost:35432" `
  --targetuser "postgres" `
  --targetpassword "YourPassword" `
  --targetdatabase "wikipedia" `
  --targetschema "public" `
  --targettable "articles" `
  --parallelmethod "DataDriven" `
  --distributeKeyColumn "(id/10000)::int" `
  --loadmode "Truncate" `
  --degree 12

DuckDB File to PostgreSQL​

Transfer from a DuckDB database file to PostgreSQL:

.\FastTransfer.exe `
  --sourceconnectiontype "duckdbstream" `
  --sourceserver "D:\OpenData\wikidata\dbpedia_14.duckdb" `
  --sourceschema "main" `
  --sourcetable "dbpedia_14" `
  --query "SELECT label, title, content, split, content_word_count, chunk, content_n_chunks, chunk_index, chunk_word_count, chunk_embedding::text FROM dbpedia_14" `
  --targetconnectiontype "pgcopy" `
  --targetserver "localhost:35432" `
  --targetuser "postgres" `
  --targetpassword "YourPassword" `
  --targetdatabase "wikipedia" `
  --targetschema "public" `
  --targettable "dbpedia_14" `
  --parallelmethod "Ntile" `
  --distributeKeyColumn "chunk_index" `
  --degree 14 `
  --loadmode "Truncate"

MSSQL to DuckDB​

Transfer from SQL Server to a DuckDB database file:

.\FastTransfer.exe `
  --sourceconnectiontype "mssql" `
  --sourceserver "localhost" `
  --sourcedatabase "tpch10_collation_bin2" `
  --sourceschema "dbo" `
  --sourcetable "orders_1M" `
  --sourcetrusted `
  --targetconnectiontype "duckdb" `
  --targetdatabase "tpch10_insert" `
  --targetschema "tpch10" `
  --targettable "orders" `
  --parallelmethod "None" `
  --loadmode "Truncate" `
  --targetserver "D:\OpenData\TPCH\data\10\duckdb\tpch10_insert.duckdb"

ClickHouse Integration​

PostgreSQL to ClickHouse - Single-threaded​

Transfer data to ClickHouse without parallelization:

.\FastTransfer.exe `
  --sourceconnectiontype "pgsql" `
  --sourceserver "localhost:15432" `
  --sourcedatabase "tpch" `
  --sourceuser "FastUser" `
  --sourcepassword "FastPassword" `
  --sourceschema "tpch_10" `
  --query "select * from tpch_10.orders limit 100000" `
  --targetconnectiontype "clickhousebulk" `
  --targetserver "localhost:8123" `
  --targetdatabase "default" `
  --targetuser "default" `
  --targetpassword "YourPassword" `
  --targetschema "default" `
  --targettable "orders" `
  --parallelmethod "None" `
  --loadmode "Append" `
  --batchsize 10000

PostgreSQL to ClickHouse - Parallel with Ctid​

High-performance transfer using PostgreSQL's Ctid method:

.\FastTransfer.exe `
  --sourceconnectiontype "pgsql" `
  --sourceserver "localhost:15432" `
  --sourcedatabase "tpch" `
  --sourceuser "FastUser" `
  --sourcepassword "FastPassword" `
  --sourceschema "tpch_10" `
  --sourcetable "orders" `
  --targetconnectiontype "clickhousebulk" `
  --targetserver "localhost:8123" `
  --targetdatabase "default" `
  --targetuser "default" `
  --targetpassword "YourPassword" `
  --targetschema "default" `
  --targettable "orders" `
  --parallelmethod "Ctid" `
  --degree 15 `
  --loadmode "Truncate" `
  --batchsize 1000000

ClickHouse to PostgreSQL - Multiple Methods​

Transfer from ClickHouse to PostgreSQL using various distribution methods:

.\FastTransfer.exe `
  --targetconnectiontype "pgcopy" `
  --targetserver "localhost:15432" `
  --targetdatabase "tpch" `
  --targetuser "FastUser" `
  --targetpassword "FastPassword" `
  --targetschema "tpch_10" `
  --targettable "orders_empty" `
  --sourceconnectiontype "clickhouse" `
  --sourceserver "localhost:8123" `
  --sourcedatabase "default" `
  --sourceuser "default" `
  --sourcepassword "YourPassword" `
  --sourceschema "default" `
  --query "select * from default.orders limit 100000" `
  --parallelmethod "None" `
  --loadmode "Append" `
  --batchsize 10000

SAP HANA Integration​

PostgreSQL to SAP HANA​

Transfer data to SAP HANA using the HANA bulk loader:

Single-threaded:

.\FastTransfer.exe `
  --sourceconnectiontype "pgsql" `
  --sourceserver "localhost:15432" `
  --sourcedatabase "tpch" `
  --sourceuser "FastUser" `
  --sourcepassword "FastPassword" `
  --sourceschema "tpch_10" `
  --sourcetable "orders" `
  --targetconnectiontype "hanabulk" `
  --targetserver "hxehost:39015" `
  --targetdatabase "" `
  --targetuser "SYSTEM" `
  --targetpassword "YourPassword" `
  --targetschema "TPCH" `
  --targettable "ORDERS" `
  --parallelmethod "None" `
  --loadmode "Append" `
  --batchsize 5000 `
  --query "select * from tpch_10.orders limit 100000"

Parallel with Ctid:

.\FastTransfer.exe `
  --sourceconnectiontype "pgsql" `
  --sourceserver "localhost:15432" `
  --sourcedatabase "tpch" `
  --sourceuser "FastUser" `
  --sourcepassword "FastPassword" `
  --sourceschema "tpch_10" `
  --sourcetable "orders" `
  --targetconnectiontype "hanabulk" `
  --targetserver "hxehost:39015" `
  --targetdatabase "" `
  --targetuser "SYSTEM" `
  --targetpassword "YourPassword" `
  --targetschema "TPCH" `
  --targettable "ORDERS" `
  --parallelmethod "Ctid" `
  --degree 4 `
  --loadmode "Truncate" `
  --batchsize 5000

SAP HANA to PostgreSQL​

Transfer from SAP HANA to PostgreSQL:

.\FastTransfer.exe `
  --targetconnectiontype "pgsql" `
  --targetserver "localhost:15432" `
  --targetdatabase "tpch_import" `
  --targetuser "FastUser" `
  --targetpassword "FastPassword" `
  --targetschema "tpch_10" `
  --targettable "orders" `
  --sourceconnectiontype "hana" `
  --sourceserver "hxehost:39015" `
  --sourcedatabase "" `
  --sourceuser "SYSTEM" `
  --sourcepassword "YourPassword" `
  --sourceschema "TPCH" `
  --sourcetable "ORDERS" `
  --parallelmethod "Random" `
  --distributeKeyColumn "o_orderkey" `
  --degree 12 `
  --loadmode "Truncate" `
  --batchsize 5000

Teradata Integration​

PostgreSQL to Teradata​

Transfer data to Teradata with parallel Ctid method:

.\FastTransfer.exe `
  --sourceconnectiontype "pgsql" `
  --sourceserver "localhost:15432" `
  --sourcedatabase "tpch" `
  --sourceuser "FastUser" `
  --sourcepassword "FastPassword" `
  --sourceschema "tpch_10" `
  --sourcetable "orders" `
  --targetconnectiontype "teradata" `
  --targetserver "192.168.239.129:1025" `
  --targetdatabase "TPCH" `
  --targetuser "UTPCH" `
  --targetpassword "UTPCH" `
  --targetschema "TPCH" `
  --targettable "orders" `
  --parallelmethod "Ctid" `
  --degree 15 `
  --loadmode "Truncate" `
  --batchsize 10000

Teradata to MSSQL​

Transfer from Teradata to SQL Server:

.\FastTransfer.exe `
  --sourceconnectiontype "teradata" `
  --sourceserver "192.168.239.129:1025"  `
  --sourcedatabase "TPCH" `
  --sourceuser "UTPCH" `
  --sourcepassword "UTPCH" `
  --sourceschema "TPCH" `
  --sourcetable "orders_15M" `
  --targetconnectiontype "msbulk" `
  --targetserver "localhost" `
  --targetdatabase "tpch10_collation_bin2" `
  --targetuser "FastUser" `
  --targetpassword "FastPassword" `
  --targetschema "dbo" `
  --targettable "orders_15M_copy" `
  --parallelmethod "RangeId" `
  --distributeKeyColumn "o_orderkey" `
  --degree 12 `
  --loadmode "Truncate"

Netezza Integration​

Transfer from Netezza to MSSQL using DATASLICEID for optimal parallelism:

.\FastTransfer.exe `
  --sourceconnectiontype "nzsql" `
  --sourceserver "NZSOURCESYS" `
  --sourcedatabase "TPCH" `
  --sourceuser "TPCH" `
  --sourcepassword "TPCH" `
  --targetconnectiontype "msbulk" `
  --targetserver "localhost" `
  --targetdatabase "tpch_test" `
  --sourceschema "tpch_10" `
  --sourcetable "orders" `
  --targettrusted `
  --targetschema "dbo" `
  --targettable "orders" `
  --parallelmethod "DataDriven" `
  --distributeKeyColumn "DATASLICEID" `
  --degree 6
Netezza DATASLICEID

Use DATASLICEID as the distribution column to leverage Netezza's internal data distribution for optimal parallel extraction.

Advanced SQL Server Techniques​

Using Partition Functions​

Leverage SQL Server partition functions for efficient parallel transfers:

.\FastTransfer.exe `
  --sourceconnectiontype "mssql" `
  --sourceserver "localhost" `
  --sourcedatabase "tpch10_collation_bin2" `
  --sourcetrusted `
  --targetconnectiontype "msbulk" `
  --targetserver "localhost" `
  --targetdatabase "tpch10_collation_bin2_copy" `
  --query "select * from orders_part" `
  --targettrusted `
  --targetschema "dbo" `
  --targettable "orders" `
  --parallelmethod "DataDriven" `
  --distributeKeyColumn "`$PARTITION.PF_DATE(o_orderdate)" `
  --degree 10 `
  --loadmode "Truncate"
Partition Function Syntax

Use `$PARTITION.FunctionName(ColumnName) to reference a partition function. The backtick (`) escapes the dollar sign in PowerShell.

Using Physloc Method​

Use physical location for parallel distribution (readonly tables only):

.\FastTransfer.exe `
  --sourceconnectiontype "mssql" `
  --sourceserver "localhost" `
  --sourcetrusted `
  --sourcedatabase "tpch_test" `
  --sourceschema "dbo" `
  --query "select * from dbo.orders_15M" `
  --targetconnectiontype "msbulk" `
  --targetserver "localhost,11433" `
  --targettrusted `
  --targetdatabase "tpch_test" `
  --targetschema "dbo" `
  --targettable "orders_ccsi" `
  --loadmode "Truncate" `
  --batchsize 120000 `
  --parallelmethod "Physloc" `
  --degree 10 `
  --mapmethod "Name"
Physloc Method

The Physloc method should only be used on readonly tables, databases, or snapshots, as it relies on physical page locations that may change during writes.

Performance Tuning​

Dynamic Degree with Negative Values​

Use negative degree values to automatically calculate parallelism based on CPU cores:

.\FastTransfer.exe `
  --sourceconnectiontype "pgsql" `
  --sourceserver "localhost:15432" `
  --sourceuser "FastUser" `
  --sourcepassword "FastPassword" `
  --sourcedatabase "tpch" `
  --sourceschema "tpch_10" `
  --sourcetable "orders" `
  --targetconnectiontype "msbulk" `
  --targetserver "localhost" `
  --targetdatabase "tpch_test" `
  --targettrusted `
  --targetschema "dbo" `
  --targettable "orders" `
  --parallelmethod "Ctid" `
  --degree -4 `
  --loadmode "Truncate"
Dynamic Degree

A negative degree value (e.g., -4) multiplies the absolute value by the number of CPU cores. With 8 cores, -4 results in 32 parallel threads.

Batch Size Optimization​

Adjust batch size based on connection type for optimal performance:

  • MSSQL (msbulk): 10,000 - 100,000 rows
  • PostgreSQL (pgcopy): 10,000 - 50,000 rows
  • Oracle (orabulk): 100,000 - 500,000 rows
  • Oracle (oradirect): 1,000 - 10,000 rows
  • ClickHouse: 10,000 - 1,000,000 rows
  • SAP HANA: 5,000 - 10,000 rows

Debug Logging​

Enable detailed logging for troubleshooting:

.\FastTransfer.exe `
  --sourceconnectiontype "pgsql" `
  --sourceserver "localhost:5432" `
  --sourceuser "FastUser" `
  --sourcepassword "FastPassword" `
  --sourcedatabase "tpch" `
  --sourceschema "tpch_10" `
  --sourcetable "lineitem" `
  --targetconnectiontype "msbulk" `
  --targetserver "localhost" `
  --targetdatabase "tpch_test" `
  --targettrusted `
  --targetschema "dbo" `
  --targettable "lineitem" `
  --parallelmethod "Ctid" `
  --degree 10 `
  --loadmode "Truncate" `
  --loglevel "debug"
Debug Mode

Use --loglevel "debug" to see detailed execution steps, connection information, and query generation details for troubleshooting.

Using Input Files​

Process SQL queries from external files:

.\FastTransfer.exe `
  --sourceconnectiontype "pgsql" `
  --sourceserver "localhost:5432" `
  --sourceuser "FastUser" `
  --sourcepassword "FastPassword" `
  --sourcedatabase "tpch" `
  --sourcefile "..\..\..\..\FastTransfert\samples\select_orders.sql" `
  --sourceschema "tpch_10" `
  --sourcetable "orders" `
  --targetconnectiontype "msbulk" `
  --targetserver "localhost" `
  --targetdatabase "tpch_test" `
  --targettrusted `
  --targetschema "dbo" `
  --targettable "orders_2" `
  --parallelmethod "Ctid" `
  --degree 6 `
  --loadmode "Truncate"

Column Mapping​

Map by Name​

Automatically map columns by name (default):

.\FastTransfer.exe `
  --sourceconnectiontype "mssql" `
  --sourceserver "localhost" `
  --sourcetrusted `
  --sourcedatabase "tpch_test" `
  --sourceschema "dbo" `
  --query "select * from dbo.orders_15M" `
  --targetconnectiontype "msbulk" `
  --targetserver "localhost,11433" `
  --targettrusted `
  --targetdatabase "tpch_test" `
  --targetschema "dbo" `
  --targettable "orders_ccsi" `
  --loadmode "Truncate" `
  --batchsize 120000 `
  --parallelmethod "Physloc" `
  --degree 10 `
  --mapmethod "Name"
tip

Build your command with the Wizard

Copyright © 2026 Architecture & Performance.