On Fri, Sep 2, 2016 at 10:32 AM, Andres Freund <andres@anarazel.de> wrote:
>
>> > I wondered if there are any figures or measurements on Postgres performance
>> > in this ‘enough memory’ environment to support or contest this point of
>> > view?
>
> I don't think that's really answerable without individual use-cases in
> mind. Answering that question for analytics, operational, ... workloads
> is going to look different, and the overheads are elsewhere.
>
> I personally think that each implementations restrictions are more
> likely to be an issue than anything "fundamental".
+1
At one point, Stonebraker was regularly claiming that "crabbing" of
buffer locks in B-Trees was a fundamental overhead paid in systems
more or less based on System R. He did eventually start to acknowledge
that Lehman and Yao figured out a technique that made that untrue in
1981, if only barely [1], but the lesson for me was to take his claims
in this area with a generous pinch of salt.
[1] https://www.cs.cmu.edu/~pavlo/static/papers/stonebraker-ic2e2014.pdf
(See his citation 11)
--
Peter Geoghegan