Re: [HACKERS] Clock with Adaptive Replacement
От | Andrey Borodin |
---|---|
Тема | Re: [HACKERS] Clock with Adaptive Replacement |
Дата | |
Msg-id | FC949382-02EE-4AAA-8D76-9513EE2470BA@yandex-team.ru обсуждение исходный текст |
Ответ на | Re: [HACKERS] Clock with Adaptive Replacement (Andres Freund <andres@anarazel.de>) |
Ответы |
Re: [HACKERS] Clock with Adaptive Replacement
Re: [HACKERS] Clock with Adaptive Replacement |
Список | pgsql-hackers |
> 24 апр. 2018 г., в 23:14, Andres Freund <andres@anarazel.de> написал(а): > > On 2018-04-24 17:16:47 +0500, Andrey Borodin wrote: >> 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. > > What problems are you seeing? This isn't a lot of code? 1. Teaching BgWriter to used data from eviction strategy to aggressively flush data to disk (instead of ++next_to_clean ) 2. Implementing strategies as lock-free algorithms for freelist These parts seem most important for benchmarking. Also: 3. Converting all rings to single buffer manager where possible 4. Using O_DIRECT while writing data files 5. Using aio and scheduling of writes These parts are not necessary, but most challenging, while not impossible though. Best regards, Andrey Borodin.
В списке pgsql-hackers по дате отправления: