I'm certain that you will find that most table-wide queries will be unbufferable. The reason is that for the most part, buffering is a result-data dependant operation. Another issue is that some standard query statements actually invoke a view, which itself is a remapping of the data in the DB.
I've used DBI and Rose::DB::Object and found that they are optimized for highly selective queries that limit the results to a RAM memory container. Rose::DB::Object::Manager abstraction will provide iterative method capabilities, but since it is build on top of DBI, it will not provide "data window" memory management.
I suggest that your applications will benefit from intelligent design rather than generic tools. I'd be very interested in any discussion that suggests otherwise because I've experienced similar design requirements.
In reply to Re: Of large database tables and high memory usage.
by snopal
in thread Of large database tables and high memory usage.
by mseabrook
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |