On Fri, Oct 28, 2011 at 2:20 PM, Merlin Moncure wrote:
>
> hrm -- it doesn't look like you are i/o bound -- postgres is
> definitely the bottleneck. taking a dump off of production is
> throwing something else out of whack which is affecting your other
> processes.
>
> band aid solutions might b
On Thu, Oct 27, 2011 at 6:29 PM, Samuel Gendler
wrote:
>
> On Thu, Oct 27, 2011 at 2:15 PM, Samuel Gendler
> wrote:
>>
>> There are definitely no bloated tables. The large tables are all
>> insert-only, and old data is aggregated up and then removed by dropping
>> whole partitions. There should
On Thu, Oct 27, 2011 at 2:15 PM, Samuel Gendler
wrote:
>
>
> There are definitely no bloated tables. The large tables are all
> insert-only, and old data is aggregated up and then removed by dropping
> whole partitions. There should be no bloat whatsoever. The OLTP side of
> things is pretty min
>From: pgsql-performance-ow...@postgresql.org
>[mailto:pgsql-performance-ow...@postgresql.org] On Behalf Of Samuel Gendler
>Sent: Thursday, October 27, 2011 12:47 PM
>To: pgsql-performance@postgresql.org
>Subject: [PERFORM] backups blocking everything
>
>I've got a large
; >To: pgsql-performance@postgresql.org
> >Subject: [PERFORM] backups blocking everything
> >
> >I've got a large mixed-used database, with the data warehouse side of
> things consisting of several tables at hundreds of millions of rows, plus a
> number of tables wi
On Thu, Oct 27, 2011 at 11:47 AM, Samuel Gendler
wrote:
> I've got a large mixed-used database, with the data warehouse side of things
> consisting of several tables at hundreds of millions of rows, plus a number
> of tables with tens of millions. There is partitioning, but as the volume
> of dat
I've got a large mixed-used database, with the data warehouse side of things
consisting of several tables at hundreds of millions of rows, plus a number
of tables with tens of millions. There is partitioning, but as the volume
of data has risen, individual partitions have gotten quite large. Hard
