Hi Folks,
I’m writing a little utility for dumping parts of tables into files which will be later slurped into another primary instance. The primary source is our referential data (big, and subject to heavy loads when adding data), the second is a smaller version used in driving our web app. (smaller, portable, less prone to lags.)
Yes, a replication strategy can work but since the web app version is so much smaller (10% of the size) I thought the partial snapshot would be easier to manage.
I have SQL that does it with \copy (select * from <table> where…) … And that is fine. But it would be nice to be able to run the \copy commands in parallel. So I was thinking of writing a background worker.
Never having done that before, I’m curious:
1) Is a background worker that I can execute in parallel appropriate for this job
2) Are there non-trivial examples of background workers out there to copy learn from?
3) Will doing multiple \copy’s in parallel just be of no benefit. Since pg_dump and pg_restore have the options of running multiple instances in parallel I thought the answer was it should help.
Thanks
Joe