On Thu, Mar 26, 2009 at 5:23 PM, Raymond O'Donnell <rod@iol.ie> wrote:
> On 26/03/2009 23:10, Scott Marlowe wrote:
>> It's also important to point out that writers don't necessarily block
>> other writers. As long as they're operating on different ranges of
>> the data set. You can have dozens of writers streaming data in with
>> differening primary keys all running together.
>
> Do you have to do anything special to have this happen - e.g. table
> partitioning? - Or does it just happen automagically based on the
> primary key?
No, it's pretty much automatic. Pgsql creates new records for every
update or insert, so there's no weird locking on the original records
to make it slow. Everything just goes in the WAL and gets flushed out
to disk later. Setting up commit siblings helps on some loads.
That's about it. No rocket science or triggers really needed for lots
of writes at the same time.