Archives of the TeradataForum

Message Posted: Sun, 01 Nov 2015 @ 10:06:28 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

I would suggest: "collect data on a non-indexed column used in joins or WHERE clause"... IF the data values in that column are very skewed / lumpy.

So if you have a table with 1M rows but only (say) three values in a column then this is going to be very 'lumpy'. The optimiser needs to know this.

So if you have a table with 1M rows with lots of values in a column but one/two values are in a large percentage of the rows, then this is going to be very 'skewed'. The optimiser needs to know this.

So if you have a table with 1M rows but lots of values and no value dominates the others then you can probably get away without stats on that column.

Cheers,

Dave

Ward Analytics Ltd - Information in motion (www.ward-analytics.com)

	<Prev	Next>		<<First	<Prev	Next>	Last>>