Hi all,

I wondering what should be the best solution for my situation. The
condition is I have to full-process a customer dimension. As a result
of full process the dimension, I have to full process a sales cube,
(haven't I?). The sales cube consists of 60 patition (60 months) and
will take about 15- 20 hours to do the full process. If I want to
reduce the processing time, what should I do?

Thanks in advance,

Anote

Assuming that the dimension is non-changing -- then you are right, a full
process of the dimension will force a full process of all cubes that use
that dimension.

If you are on a low-end single or dual CPU system, then use DTS and SQL
Agent to automate your processing. A simple DTS package with two OLAP
Processing task steps would do it -- one for the dimension and one for the
cube. If you process the cube, then each of the partitions are processed
sequentially.

If running on a 4X or larger multiprocessor server, look at the Parallel
Processing Utility.
http://www.microsoft.com/downloads/d...DisplayLang=en
You can set a workload to process the dimension and then it will process
each partition in parallel. It can be launched from a command line and
scheduled via SQL Agent.
--
Dave Wickert [MS]
dwickert (AT) online (DOT) microsoft.com
Program Manager
BI Practices Team
SQL BI Product Unit (Analysis Services)
--
This posting is provided "AS IS" with no warranties, and confers no rights.


"Anote Sangkunakupt" <anote_un (AT) hotmail (DOT) com> wrote