Loading Data with COPY

A newer version of this documentation is available. Click here to view the most up-to-date release of the Greenplum 4.x documentation.

Loading Data with COPY

COPY FROM copies data from a file or standard input into a table and appends the data to the table contents. COPY is non-parallel: data is loaded in a single process using the Greenplum master instance.

To optimize the performance and throughput of COPY, run multiple COPY commands concurrently in separate sessions and divide the data evenly across all concurrent processes. To optimize throughput, run one concurrent COPY operation per CPU.

The COPY source file must be accessible to the master host. Specify the COPY source file name relative to the master host location.

Greenplum copies data from STDIN or STDOUT using the connection between the client and the master server.