Ühel kenal päeval, T, 2007-12-11 kell 13:44, kirjutas Csaba Nagy:
> On Tue, 2007-12-11 at 11:12 +0000, Simon Riggs wrote:
> > Features
> > - Read Only Tables
> > - Compressed Tablespaces
>
> I wonder if instead of read-only tables wouldn't it be better to have
> some kind of automatic partitioning which permits to have different
> chunks of the table data in different tablespaces, and a freeze command
> which effectively moves the data from the (normally small) active chunk
> to the archive chunk when it's transaction id is older than a predefined
> threshold ?
This would be doable using Simons proposed commands.
> Then put the active chunk on a high performance file system and the
> archive tablespace on a compressed/slow/cheap file system and you're
> done. Allow even the archive chunk to be updateable, and put new tuple
> data in the active chunk. It would work just fine for cases where the
> old data is rarely updated/deleted...
You can't update a table on a read-only (write-once) partition, at least
not with current header structure.
> Another advantage I guess would be that active data would more likely
> stay in cache, as updated records would stay together and not spread
> over the inactive.