[Info-Ingres] Automatic partitioninng and unique keyed tables

Thread view:

Enable EMail Alerts

Start New Thread

Thread rating:

martin.bowes@ctsu.ox.ac.uk - 16 Apr 2007 14:54 GMT

Hi Everyone,

I'm running Ingres2006 +p12125 on a64.lnx

Why is it that I cannot put an automatic partition on a unique keyed
table?

As I understand it, the automatic partition is just there to scatter the
data randomly amongst the partitions. So to retrieve data it must go and
look in each partition. Surely uniqueness would actually help not hinder
that task?

Martin Bowes
--
Random Farscape Quote #3:
John - If this is part of some sick experiment, I'm going to bill them for
the therapy.

Reply to this Message

Roy Hann - 16 Apr 2007 15:13 GMT

> Hi Everyone,
>
[quoted text clipped - 7 lines]
> look in each partition. Surely uniqueness would actually help not hinder
> that task?

I could wait to let Karl tell us all for sure, but my supposition has always
been that in general, the cost of verifying uniqueness when the confounding
value could (in principle) appear in one any of 65,000 partitions is just
too high. Ingres is protecting you from doing something expensive.

My rule of thumb is that if the parser couldn't figure out where to look,
partitioning will just make things slower.

Roy

Reply to this Message