Horton

Re: POC: postgres_fdw insert batching

Andrei Lepikhov <a.lepikhov@postgrespro.ru>

View thread View on postgr.es
From: "Andrey V. Lepikhov" <a.lepikhov@postgrespro.ru>
To: Tomas Vondra <tomas.vondra@2ndquadrant.com>, pgsql-hackers@postgresql.org
Date: 2020-07-10T04:28:44Z
Lists: pgsql-hackers 
On 6/28/20 8:10 PM, Tomas Vondra wrote:
> Now, the primary reason why the performance degrades like this is that
> while FDW has batching for SELECT queries (i.e. we read larger chunks of
> data from the cursors), we don't have that for INSERTs (or other DML).
> Every time you insert a row, it has to go all the way down into the
> partition synchronously.

You added new fields into the PgFdwModifyState struct. Why you didn't 
reused ResultRelInfo::ri_CopyMultiInsertBuffer field and 
CopyMultiInsertBuffer machinery as storage for incoming tuples?

-- 
regards,
Andrey Lepikhov
Postgres Professional

Commits

  1. Fix tuple routing to initialize batching only for inserts

  2. Fix pointer type in ExecForeignBatchInsert SGML docs

  3. Fix initialization of FDW batching in ExecInitModifyTable

  4. Implement support for bulk inserts in postgres_fdw