it's highly unclear what such a "study" would present: there are too many variables in the model you chose, as well as other related on what db engine you are using. Let's say we do a "posts --> tags" model.
I can get home and present you a mockup model tested on postgresql, with timings on updates/writes/reads on both models (lists:reference or straight joins) but that's going to be valid on my machine, with my db config, my ram and my cpu, my model and my data cardinality (i.e. how many posts and how many tags). Normalization and de-normalization have their own pros and cons (and a long list of books have been written), and they're related both on the "target" system (i.e. you need an app that just stores tags and searches through them or you need an app where tags gets managed too?) and on the underlying db technology. On Thursday, February 28, 2013 2:23:55 PM UTC+1, Cliff Kachinske wrote: > > Has anyone studied the effect on database performance when using > list:reference fields in a relational database? > > If so, could you share the data? > > Thanks, > Cliff Kachinske > -- --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
