After working with Access & MySQL, I discovered several functionalities that weren't possible with just Access, and I know that from my past search (may not be true anymore) while researching various RDBMSs, that there wasn't anything comprehensive discussing what functionalities became available/possible when using a given Access/RDBMS combination. So, I thought it would be interesting to have a list showing what Access could do with a given backend and hopefully this may benefit anyone who has luxury of choosing a RDBMS (which I assume is not true in most of cases??) in making a informed decision or at least provide food for thought (that'd be me!).
And if you're scratching head wondering what the heck 'functionality' Banana is talking about, let me cite two examples with MySQL.
One, I've already stated earlier that I can have much more control over transactions, especially across bound form/subforms, committing or rolling back *both* parent record and related child records as whole. This isn't possible with Access backend, AFAIK.
Two, MySQL has one pluggable storage engine, InnoDB which enables me to read records without actually requiring locks. This is possible using a form of version control so reads never ever locks the row unless explicitly requested by the client. This is a good thing to have if deadlock is a concern. (To be fair, I'm not sure if that is also the case with other RDBMS- do please correct me!)
So, I know some of you use Oracle, and more use MS SQL... I don't know anyone else here uses PostgreSQL, FireBird, or ??? Maybe they can chime in, too.
And if you're scratching head wondering what the heck 'functionality' Banana is talking about, let me cite two examples with MySQL.
One, I've already stated earlier that I can have much more control over transactions, especially across bound form/subforms, committing or rolling back *both* parent record and related child records as whole. This isn't possible with Access backend, AFAIK.
Two, MySQL has one pluggable storage engine, InnoDB which enables me to read records without actually requiring locks. This is possible using a form of version control so reads never ever locks the row unless explicitly requested by the client. This is a good thing to have if deadlock is a concern. (To be fair, I'm not sure if that is also the case with other RDBMS- do please correct me!)
So, I know some of you use Oracle, and more use MS SQL... I don't know anyone else here uses PostgreSQL, FireBird, or ??? Maybe they can chime in, too.