On 10/16/2010 4:07 AM, Thomas wrote:

>> My specific question is whether any system actually collects
>> potential conflicts and analyzes them at commit time rather
>> than recognizing the conflict on update and resolving it by
>> requiring the second concurrent transaction trying to update
>> or delete a row to wait for the outcome of the first transaction?
>
> Not sure I understand the question completely, but wouldn't a
> DBMS that allows deferred constraints need to do exactly that?

Actually, not exactly. The better way to handle deferred constraints
is to perform the constraint check immediately, and retain only the
failures, then recheck the failures on commit. Some - in a properly
designed system all - of the failures will be corrected. There's
no reason to recheck all the actions of the transaction.


Best regards,


Ann