Re: Clock sweep not caching enough B-Tree leaf pages?

On Wed, Apr 16, 2014 at 1:42 PM, Peter Geoghegan <pg(at)heroku(dot)com> wrote:
> On Wed, Apr 16, 2014 at 4:01 AM, Andres Freund <andres(at)2ndquadrant(dot)com> wrote:
>>> Aren't you interested in the significance of the patch, and the test case?
>>
>> Not particularly in the specifics to be honest. The tradeoffs of the
>> techniques you used in there seem prohibitive to me. It's easy to make
>> individual cases faster by sacrificing others.
>
> You're the one poring over the specifics of what I've done, to my
> consternation. I am not prepared to defend the patch at that level, as
> I've made abundantly clear. I've called it a sketch, a proof of
> concept half a dozen times already. I don't understand your difficulty
> with that. I also don't understand how you can be so dismissive of the
> benchmark, given the numbers involved. You're being unreasonable.

I don't think he's being unreasonable, and I don't understand why
you're getting bent out of shape about it. You proposed a patch, he
articulated a problem, you don't want to fix it right now. All of
which is fine. Why the ad hominem accusations?

> If I didn't write this patch, and I talked to people about this issue
> at pgCon, I'm not sure that anyone would be convinced that it was a
> problem, or at least that it was this much of a problem.

I agree with that, too.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company