Archives of the TeradataForum
Message Posted: Wed, 28 Apr 2004 @ 13:57:26 GMT
| Subj: || || Re: Teradata Data Type Percentages |
| From: || || Geoffrey Rommel |
| ||Actually all I need is to find out how I can get the size of all the indexes in our system. From What I was told, primary indexes do not
take much space, if any.|| |
That's one way of looking at it, yes. Every table must have a primary index (which is really just a partitioning key), so the index cannot be
dropped. Nevertheless, smaller primary indexes are better, since the actual data value is stored along with the row hash.
| ||Does anyone know of a trick to come up with this? maybe a script against the data dictionary?|| |
Unfortunately, the data dictionary does not contain information on the size of secondary indexes. It really should, IMHO. The only way to find
the size of the secondary indexes and other subtables is to run a utility from a TPA node (ferret, I believe).