On Thu, 2007-08-02 at 18:06 +0100, Csaba Nagy wrote:
> The problem is that eliminating the deadlock is still not the complete
> cake... the interlocking still remains, possibly leading to degraded
> performance on high contention on very common parent rows. The real
> solution would be when an update on the parent table's non-referenced
> fields is not interlocking at all with updates of the child rows... and
> I think there were some proposals to do that.
Agreed. There are two issues here, unnecessary blocking and deadlock.
These can be tackled separately. My proposal deals only with the
deadlock issue.
Even if if contention is reduced, for instance by implementing
column-level locking, there will still be the potential for deadlock
arising from inconsistent ordering of locks. I continue to stand by my
proposal.
__
Marc