> What about some C dynamically loaded function in which I could call new connection for each thread?
Past discussion here suggests that the backends are strictly single
threaded. While you might be able to use multiple threads - I don't know
- I expect you'd need to protect all SPI access by a lock that
serialized everything anyway.
Doing it externally with a script / program that uses multiple
connections might just be the way. Unfortunately that means that you
don't get a single consistent snapshot - each connection will have its
own, potentially different, view of the database state.
A possible use for read only transactions being able to share a snapshot
came up in discussion a few weeks ago. I guess this is another one.
In response to
pgsql-general by date
|Next:||From: A. Kretschmer||Date: 2008-03-30 17:29:22|
|Subject: Re: Re: [GENERAL] postgreSQL multithreading|
|Previous:||From: Craig Ringer||Date: 2008-03-30 17:17:10|
|Subject: Re: Survey: renaming/removing script binaries (createdb,