batchsize setting sets the data batch loading size.
This is a session setting.
By default, the batchsize is 5000.
Too large of batch size consumes too much memory, while too small of batch size can significantly reduce the loading performance. A good value is the one that gets the most out of the loading performance without consuming a lot of memory.
Depending on the row size (i.e. how big a single row of data to be transmitted in bytes), 5000 is a fairly good value. On extremely large rows, you may want to reduce this value.
usage: .set batchsize [size]
Without the size arguments, it displays the current session’s batch size.
If the batch size is 1, batch mode is disabled.
-- Disable batch mode .set batchsize 1 .import csv -h on batch.csv INSERT INTO MyTable VALUES (?, ?); -- Large batch .set batchsize 10000 .import csv -h on batch.csv INSERT INTO MyTable VALUES (?, ?);