hasta_l3_at_hotmail.com a écrit :
> >
> > I am led, by what you've written, to question whether you understand
> > basic Oracle architecture and concepts. I still don't quite see this
> > as being anything more than overcomplicating something unnecessarily.
> >
> > Oracle is not SQL Server. Why is it that any locking is required?
>
> I never used SQL Server, Daniel.
>
> To recap the context :
>
> > Assume a (long) restructuring operation L modifies
> > (a) some fields of the list header row in the master table, then
> > (b) some fields of every list line row in the children table, including
> > -say - row R.
> >
> > Assume that a short transaction S modifies
> > (a) some other fields of the list line row R.
> >
> > If L is in a single long transaction, then it will block S if S happens
> > to want to modify R after L did it.
>
> Right now, that problem is solved by breaking the long operation
> in many small transactions.
>
> I'm just enumerating ways to replace the burst of thousands of
> transactions by a single long one, if profiling shows it to be
> necessary.
And also because the "natural" size of the long operation is a
single transaction, and that would simplify some peripheral issues.
Received on Sat Nov 04 2006 - 14:14:38 CST
