Archives of the TeradataForum

Message Posted: Mon, 11 Jul 2011 @ 13:12:33 GMT

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

Thanks a lot Dave. This is very good.

Can you please confirm my understanding:

With Fastload, In first phase rows from client system directly go to the target table on correct amps, but are unsorted.

Since they go to the target table , the target table is not accessible during this phase.

Later END LOADING statement triggers the phase 2 when the rows on correct AMPs are sorted as per Rowids(or rowkeys) and duplicates are removed.

With multiload , in acq phase the rows from client system go to the worktable on correct AMPs.

Since target table is not affected in this phase, it remains accessible in multiload phase.

And in application phase rows are inserted/updated to the target table from the work table.

Please let me know if my understanding is correct.

Questions:

1) Does that mean that fastload does not need any spool during the phase 1.?

2) Does multiload involve sorting as in case of fastload.

Thanks in advance.

Sukul

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