Posted by Tim on 06/25/07 09:50

Hi Folks,

I'm not certain if this is the correct group to post this question in,
if there is a more appropriate one please advise.

We have a cube that takes a little time to build, length of time
varies by point in day and source SQL Server loading. While building
the cube is 'off-line' and this is bothering the powers that be. As
the underlying SQL Server data warehouse grows the time 'off-line' is
bound to lengthen.
..
This cube is currently being built twice a day but a more frequent
schedule of builds is envisaged.

I have been tasked to investigate options, (if any of you wise folks
have printable suggestions please post).

One option that has perculated it's way to the top of the corporate
consciousness is to build a duplicate cube in parallel to the original
so that the original can stay on-line during the build. Then once the
duplicate is fully built, drop the original and re-name the duplicate
to take the originals place?

RTFM on this suggestion doesn't get me very far and I am beginning to
think that re-naming a cube is functionality that is not available?

To make life even more fun I'm not a 'Cube' person, I'm just the data
bod who as been given the task while the cube person's on leave.

Any help, steers, wisedom, etc.. gratefully accepted

TIA Tim

• Next in forum: problem with highscore
• Prev in forum: SQL 2000 date format problem after migration W2k to W2k3
• Thread view: Parallel build and subsequent re-name of a cube

[Reply to this message]