on 06-27-2005 10:26 AM
Hi,
How can I evaluate the workload to install and setup new data flows for QM in BW?
I've seen in BCT (BW3.5) that there are more than 10 InfoCubes.
What rule can I use to make a gross estimation? Based on the number of Cubes?
Thxs.
LauQ
If you are concerned about your system capacity, perhaps your DBA could help you. But if your concern is manhours, you could always run the install in the background!
The run time lenghts of records will give you an idea of the size of each record in bytes. But you can't do this until you install it!
And then the number of records depends on the transactions in R/3 or whatever system you are using for QM activities.
Message was edited by: Sundar Raman
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi Laurent,
Speaking from my experience for PP,MM,SD, and FI/CO (unfortunately, we did not implement PM and only QM lite)it should take about 2 days to install the BC for PM/QM <i>only on the dev server</i>.
When we started up (from scratch), our disk space for Dev and QA systems were shared and was of the order of a 100 GB or so, but did not use up all of it.
I still don't get what you mean by "workload".
You start at the most atomic level in BW, that is, the InfoObject Catalog, then InfoProviders, then InfoSource and Comm. Strucutres, then Transfer Structures & Rules, then Update Rules.
Sorry, I think I would question the 3-4 week estimates and the no customization assumption when you are talking about full production go-live type scenerio.
I have not worked with QM and PM in 3.5 but those are highly customized functional areas such that QM and PM would be different for a manufacturing plant versus an utility versus a transportation company. In past versions, PM and QM standard content was not sufficient out of the box and required significant customization.
Repeat, this is not like FI-CO where BCT can cover 80-90% of your needs out of the box and where extensions can be done easily and the datasources / cubes sufficiently tested by other projects that they are no major issue. I think a 3-4 week assumption for timeline to golive is underestimation and leaves the project in the high risk zone. 3-4 weeks might be possible in some circumstances but I think the chance of the project not achieving the real objectives is extremely high.
Just my 2 cents
User | Count |
---|---|
91 | |
10 | |
10 | |
9 | |
9 | |
7 | |
6 | |
5 | |
5 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.