Re^4: Efficient way to handle huge number of records?

I don't know your PRAGMA code, but in general this is a possible solution.

The only thing where I'd like to comment is that DBI allows and you should use a $dbh->begin_work and a $dbh->commit instead of $dbh->do( "BEGIN" ) and $dbh->do( "COMMIT" );

Comment on Re^4: Efficient way to handle huge number of records? Select or Download Code

Replies are listed 'Best First'.
Re^5: Efficient way to handle huge number of records? by Marshall (Canon) on Dec 11, 2011 at 14:51 UTC
I think that your formulation of "begin transaction" and "commit transaction" is more general and therefore better! The PRAGMA stuff is specific to SQLite. It is possible to dynamically increase the size of the memory cache that SQLite uses and actually to decrease it later. I did it to speed up the indexing operation which is actually quite complex and builds a big memory structure to do it.	[reply]