Re: [REVIEW] Re: Compression of full-page-writes
| От | Andres Freund |
|---|---|
| Тема | Re: [REVIEW] Re: Compression of full-page-writes |
| Дата | |
| Msg-id | 20140929152911.GM16581@awork2.anarazel.de обсуждение |
| Ответ на | Re: [REVIEW] Re: Compression of full-page-writes (Heikki Linnakangas <hlinnakangas@vmware.com>) |
| Список | pgsql-hackers |
On 2014-09-29 18:27:01 +0300, Heikki Linnakangas wrote: > On 09/29/2014 06:02 PM, Robert Haas wrote: > >On Mon, Sep 29, 2014 at 8:36 AM, Andres Freund <andres@anarazel.de> wrote: > >>1) I don't think it's a good idea to put the full page write compression > >> into struct XLogRecord. > > > >Why not, and where should that be put? > > It should be a flag in BkpBlock. Doesn't work with the current approach (which I don't really like much). The backup blocks are all compressed together. *Including* all the struct BkpBlocks. Then the field in struct XLogRecord is used to decide whether to decompress the whole thing or to take it verbatim. Greetings, Andres Freund -- Andres Freund http://www.2ndQuadrant.com/PostgreSQL Development, 24x7 Support, Training & Services
В списке pgsql-hackers по дате отправления: