On Mon, Dec 21, 2015 at 7:51 AM, Fabien COELHO <coelho@cri.ensmp.fr> wrote:
> I think that the 1.5 value somewhere in the patch is much too high for the
> purpose because it shifts the checkpoint load quite a lot (50% more load at
> the end of the checkpoint) just for the purpose of avoiding a spike which
> lasts a few seconds (I think) at the beginning. A much smaller value should
> be used (1.0 <= factor < 1.1), as it would be much less disruptive and would
> probably avoid the issue just the same. I recommend not to commit with a 1.5
> factor in any case.
Wait, what? On what workload does the FPW spike last only a few
seconds? That's certainly not the case in testing I've done. It
would have to be the case that almost all the writes were concentrated
on a very few pages.
> Another issue I raised is that the load change occurs both with xlog and
> time triggered checkpoints, and I'm sure it should be applied in both case.
Is this sentence missing a "not"?
> Another issue is that the patch makes sense when the WAL & relations are on
> the same disk, but might degrade performance otherwise.
Yes, that would be a good case to test.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company