On Apr 5, 2005 12:16 AM, Christopher Petrilli <petrilli@gmail.com> wrote:
> Looking at preliminary results from running with shared_buffers at
> 16000, it seems this may be correct. Performance was flatter for a
> BIT longer, but slammed right into the wall and started hitting the
> 3-30 second range per COPY. I've restarted the run, with fsync turned
> on (fdatasync), and we'll see.
>
> My fear is that it's some bizarre situation interacting with both
> issues, and one that might not be solvable. Does anyone else have
> much experience with this sort of sustained COPY?
Well, here's the results:
http://www.amber.org/~petrilli/diagrams/pgsql_copy500_comparison.png
The red is the run with shared_buffers turned up, but fsync off.
The blue is the run with shared_buffers turned up, but fsync on.
Note that it hits the wall sooner. Unfortunately, my brain is fried,
and not sure what that means!
Chris
--
| Christopher Petrilli
| petrilli@gmail.com
