On Fri, 5 Feb 1999, Tom Lane wrote:
> [ Redirected to hackers, as it's now way off-topic for pgsql-interfaces ]
>
> Peter T Mount <peter@retep.org.uk> writes:
> > [ trying to test whether Postgres can split a huge database into
> > multiple 2-Gb files ]
> > The problem I have is that it takes 4 hours for a table to reach 2Gb on my
> > system, so it's a slow process :-(
>
> I had a thought about this --- is there a #define somewhere that sets
> the size at which the system decides it needs to split a table?
> (If not, shouldn't there be?) If there is, you could build a debug
> copy of Postgres in which the value is set very small, maybe a meg or
> less, and then it'd be easy to test the file-splitting behavior.
I've already found it, and I was thinking along those lines.
The define was last changed when we got variable block sizes, and there's
some lengthy comments by it describing how it's value's calculated.
Anyhow, I'm planning on working on this tomorrow morning.
Peter
-- Peter T Mount peter@retep.org.uk Main Homepage: http://www.retep.org.uk
PostgreSQL JDBC Faq: http://www.retep.org.uk/postgresJava PDF Generator: http://www.retep.org.uk/pdf