Thomas Kellerer wrote:
> I don't think we have a caching issue though, it's just one (regular)
> bulk import that does some heavy writes which we need to tune. Read
> performance is absolutely fine, but everything around vacuum and
> checkpointing seems to slow down the import massively.
> So would taking a snapshot before and after this import be the right
> way to start the analysis?
Sure. Might want to grab one at those two points, then at a point right
before the next big import happens, so that you have what a relatively
idle period looks like for comparison sake.
--
Greg Smith 2ndQuadrant US Baltimore, MD
PostgreSQL Training, Services and Support
greg@2ndQuadrant.com www.2ndQuadrant.us