Sudipta,
The beautiful thing is _why_ primary index doesn't cost you space. This is because Primary Index is effectively a part of the data storage mechanism that addresses Data blocks on the AMPs: Master Index -> Cylinder Index -> Data block. The details are available in Teradata documentation. As a result, you get two things at once - ability to address certain data blocks, and ability to locate data block(s) that correspond to certain Primary Index value (of a certain table). Therefore, you don't have to pay extra perm space to get the index access for the Primary Index.
Nishant,
you can use dbc.tablesize view to see the size of table plus its secondary indices. Create a table without secondary index. Note the size. Then create your new secondary index. Note the new size. The difference is the size of the index.
There are other methods to do that, but this one is the most straightforward one.
Yes, you should take Secondary Indices into account during Capacity planning.
Regards,
Vlad.
Sudipta,
The beautiful thing is _why_ primary index doesn't cost you space. This is because Primary Index is effectively a part of the data storage mechanism that addresses Data blocks on the AMPs: Master Index -> Cylinder Index -> Data block. The details are available in Teradata documentation. As a result, you get two things at once - ability to address certain data blocks, and ability to locate data block(s) that correspond to certain Primary Index value (of a certain table). Therefore, you don't have to pay extra perm space to get the index access for the Primary Index.
Nishant,
you can use dbc.tablesize view to see the size of table plus its secondary indices. Create a table without secondary index. Note the size. Then create your new secondary index. Note the new size. The difference is the size of the index.
There are other methods to do that, but this one is the most straightforward one.
Yes, you should take Secondary Indices into account during Capacity planning.
Regards,
Vlad.