Re: Raising the checkpoint_timeout limit

On 2/2/16 10:10 PM, Robert Haas wrote:
> Now, you could also set such configuration settings in
> a situation where it will not work out well.  But that is true of most
> configuration settings.

Yeah, if we're going to start playing parent then I think the first 
thing to do is remove the fsync GUC. The AWS team has done testing that 
shows it to be worthless from a performance standpoint now that we have 
synchronous commit, and it's an extremely large foot-bazooka to have 
laying around.
-- 
Jim Nasby, Data Architect, Blue Treble Consulting, Austin TX
Experts in Analytics, Data Architecture and PostgreSQL
Data in Trouble? Get it in Treble! http://BlueTreble.com