Re: Prototype: In-place upgrade v02
От | Zdenek Kotala |
---|---|
Тема | Re: Prototype: In-place upgrade v02 |
Дата | |
Msg-id | 48C4DDF9.9050507@sun.com обсуждение исходный текст |
Ответ на | Re: Prototype: In-place upgrade v02 (Bruce Momjian <bruce@momjian.us>) |
Список | pgsql-hackers |
Bruce Momjian napsal(a): > > As far as the page not fitting after conversion, what about some user > command that will convert an entire table to the new format if page > expansion fails. Keep in a mind that there are more kind of pages. Heap is easy, but each index AM has own specific :(. Better approach is move tuple to the new page and invalidate all related table indexes. Following reindex automatically convert whole table. >> After putting all those together, how large a patch are we talking >> about, and what's the performance penalty then? How much of all that >> needs to be in core, and how much can live in a pgfoundry project or an >> extra binary in src/bin or contrib? I realize that none of us have a >> crystal ball, and one has to start somewhere, but I feel uneasy >> committing to an approach until we have a full plan. > > Yes, another very good point. > > I am ready to focus on these issues for 8.4; all this needs to be > fleshed out, perhaps on a wiki. As a starting point, what would be > really nice is to start a wiki that lists all data format changes for > every major release. As Greg mentioned in his mail there wiki page is already there. Unfortunately, I did not time to put actual information there. I'm going to do soon.Zdenek -- Zdenek Kotala Sun Microsystems Prague, Czech Republic http://sun.com/postgresql
В списке pgsql-hackers по дате отправления: