On Tue, Nov 30, 2004 at 10:57:02AM -0500, Tom Lane wrote:
> Greg Stark <gsstark@mit.edu> writes:
> > Mark Wong <markw@osdl.org> writes:
> >> I have some initial results using 8.0beta5 with our OLTP workload.
> >> http://www.osdl.org/projects/dbt2dev/results/dev4-010/199/
> >> throughput: 4076.97
>
> > Do people really only look at the "throughput" numbers? Looking at those
> > graphs it seems that while most of the OLTP transactions are fulfilled in
> > subpar response times, there are still significant numbers that take as much
> > as 30s to fulfil.
>
> > Is this just a consequence of the type of queries being tested and the data
> > distribution? Or is Postgres handling queries that could run consistently fast
> > but for some reason generating large latencies sometimes?
>
> Given the regular shape of the first graph (transactions/minute),
> there's every reason to think that the slowdowns are caused by
> checkpoint I/O storms. It would be worth the trouble to experiment
> with adjusting the bgwriter parameters to even out the flow of write
> operations. (I think I've already pointed out that the current defaults
> for the bgwriter seem mighty conservative.)
>
> regards, tom lane
I do have bgwriter_delay increased to 10, per previous
recommendation, which did smooth out the throughput graph
considerably. I can continue to adjust those settings.
Mark
