Re: [HACKERS] Clock with Adaptive Replacement

вт, 24 апр. 2018 г., 15:16 Andrey Borodin <x4mmm(at)yandex-team(dot)ru>:

>
>
> > 24 апр. 2018 г., в 11:31, Юрий Соколов <funny(dot)falcon(at)gmail(dot)com>
> написал(а):
> >
> > Before implementing algorithms within PostgreSQL it will be great to
> test them outside with real traces.
> >
> > I think, there should be mechamism to collect traces from real-world
> postgresql instalations: every read and write access. It should be
> extremely eficient to be enabled in real world. Something like circular
> buffer in shared memory, and separate worker to dump it to disk.
> > Instead of full block address, 64bit hash could be used. Even 63bit +
> 1bit to designate read/write access.
> Yes, this is good idea to track pattern of blocks usage.
> But, I think that cost of development of real page eviction strategy
> itself is neglectable small compared to infrastructure changes needed by
> any non-CS5 strategy.
>
> Best regards, Andrey Borodin.

Exactly. And it will be pity if implemented strategy will not be "the best".

Different strategies may need different infrastructure. And implementing
five infrastructures are certainly more expensive than choosing two best
strategies using traces and implementing them.