Re: Refactoring of heapam code.
От | Anastasia Lubennikova |
---|---|
Тема | Re: Refactoring of heapam code. |
Дата | |
Msg-id | 87f83470-0f54-ffa6-3635-51c0f44c8f72@postgrespro.ru обсуждение исходный текст |
Ответ на | Re: Refactoring of heapam code. (Kevin Grittner <kgrittn@gmail.com>) |
Список | pgsql-hackers |
05.08.2016 16:30, Kevin Grittner: > On Fri, Aug 5, 2016 at 2:54 AM, Anastasia Lubennikova > <a.lubennikova@postgrespro.ru> wrote: > >> They can be logically separated into three categories: >> "primary storage" - r, S, t, v. They store data and visibility information. >> The only implementation is heapam.c >> "secondary index" - i. They store some data and pointers to primary storage. >> Various existing AMs and managed by AM API. >> "virtual relations" - c, f, m. They have no physical storage, only entries >> in caches and catalogs. > Materialized views (relkind == "m") have physical storage, and may have indexes. Good point. I сonfused letters for views (virtual relations) and materialized views (primary storage). Should be: "primary storage" - r, S, t, m. They store data and visibility information. The only implementation is heapam.c "secondary index" - i. They store some data and pointers to primary storage. Various existing AMs and managed by AM API. "virtual relations" - c, f, v. They have no physical storage, only entries in caches and catalogs. -- Anastasia Lubennikova Postgres Professional: http://www.postgrespro.com The Russian Postgres Company
В списке pgsql-hackers по дате отправления: