It's been pulled to close-source by Broadcom now, but if anyone is interested there was a fairly cool from-scratch go implementation of the same basic ideologic

It's been pulled to close-source by Broadcom now, but if anyone is interested there was a fairly cool from-scratch go implementation of the same basic ideological approach as pg_dump for Greenplum. Fairly cool for seeing the catalog queries it takes to pull each piece of information that goes into a schema, at the very least. https://github.com/greenplum-db/gpbackup-archive https://github.com/greenplum-db/gpbackup-archive