Archives of the TeradataForum

Message Posted: Wed, 25 May 2005 @ 13:57:24 GMT

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

From how you describe the problem I think it must be related to the data skew, and probably the new data is responsible for it.

You can always inverstigate what exactly is wrong with the data but it might take some time.

If you need to fix the problem quickly, change the Primary Index definition on the empty (staging) table you are loading into. How? I don't know because I have not seen the table definition but usually adding some additional column with high cardinality of data values would do. Perhaps some timestamp if it exists there?

That would ensure a more even distribution of the data in the staging area and you have a chance to load the data at least to that table.

That might not help to finish the whole loading cycle as the target table might have the same data skew issue but if you have the data in the staging table you can at least easily count the number of records per PI value (current defintition). You should easily find the outlier.

Hope this helps.

Victor

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