This section describes the formula for estimating memory requirements for the high-speed data load feature.
Use the following formula to obtain a rough estimate of memory requirements:
Memory usage of high speed data load = (Peak memory usage of pgx_loader processes + Peak memory usage of the pgx_loader commands) x Number of commands executed simultaneously Peak memory usage of pgx_loader processes = Peak memory usage of the backend process (6 MB) + Peak memory usage of parallel workers (6 MB x number of parallel workers) + Peak memory usage of dynamic shared memory (80 MB x number of parallel workers) Peak memory usage of the pgx_loader commands=9 MB
Point
In addition to the size calculated using the formula above, the database cache on the shared memory estimated using the shared_buffers parameter is consumed according to the size of the data (table and index keys) loaded using this feature. Refer to "E.1 Estimating Table Size Requirements" and "E.2 Estimating Index Size Requirements" for information on estimating an appropriate shared buffers value.