Re: Speeding up INSERTs and UPDATEs to partitioned tables

From: Jesper Pedersen <jesper(dot)pedersen(at)redhat(dot)com>
To: David Rowley <david(dot)rowley(at)2ndquadrant(dot)com>, Amit Langote <Langote_Amit_f8(at)lab(dot)ntt(dot)co(dot)jp>
Cc: Robert Haas <robertmhaas(at)gmail(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Speeding up INSERTs and UPDATEs to partitioned tables
Date: 2018-11-06 18:56:17
Message-ID: 06524959-fda8-cff9-6151-728901897b79@redhat.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 11/4/18 5:07 AM, David Rowley wrote:
> I've attached v13 which hopefully addresses these.
>

I ran a test against the INSERT case using a 64 hash partition.

Non-partitioned table: ~73k TPS
Master: ~25k TPS
0001: ~26k TPS
0001 + 0002: ~68k TPS

The profile of 0001 shows that almost all of
ExecSetupPartitionTupleRouting() is find_all_inheritors(), hence the
last test with a rebase of the original v1-0002 patch.

Best regards,
Jesper

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Robert Haas 2018-11-06 18:56:37 Re: ATTACH/DETACH PARTITION CONCURRENTLY
Previous Message Simon Riggs 2018-11-06 18:54:35 Re: ATTACH/DETACH PARTITION CONCURRENTLY