Hi, I have a question. I want to use a data warehouse to keep the books from a library. From what I've read I know that a data warehouse keeps historical data. I would like to keep informations about a library, and make queries. I would also like to keep a book for a certain costumer, so i would have to update some data. If I use a cube, can I update it? If I can, do I have to reprocess it? Please tell me, is this the best solution for my project?
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations strongm on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
Update cube
- Thread starter Goodya
- Start date
Hi,
I have worked with COGNOS cubes.
At the time of cretion, a cube retrieves data from the database & keeps it within itself.
After that, there is no connectivity between cube & database. It functions independently.
If there is any data appended to the Database, the cube has to be updated either fully or incrementally.
Prasad
RPrasad1@Chn.Cognizant.com
I have worked with COGNOS cubes.
At the time of cretion, a cube retrieves data from the database & keeps it within itself.
After that, there is no connectivity between cube & database. It functions independently.
If there is any data appended to the Database, the cube has to be updated either fully or incrementally.
Prasad
RPrasad1@Chn.Cognizant.com
From an MS Analysis Services point of view you'd be best creating the cube from your original databases data. You can then choose where you store the cubes data and aggregations and (as above) whenever you change the source data you can tell the cube to reprocess (which essentially will re-grab the data from the source database). I have mine set-up to fully re-process every morning before I come in but there's only a few million rows in it at the moment - will have to set up incremental when I put the rest of the data in.
If you wanted to leave the database alone then - once you've built the cube in Analysis Manager - you can enable the write-back feature. This means that you can actually edit your base data (in this case probably you're library database) from whatever you're viewing the cubes with. Personally I've never played with this part but I know people who have - they've allowed people to adjust their reports' targets through the cube rather than build them an interface into the database.
Does any of that make sense? Drawback to working late is that minds gradually fade - you should see some of the SQL I just tried to run
Mike
If you wanted to leave the database alone then - once you've built the cube in Analysis Manager - you can enable the write-back feature. This means that you can actually edit your base data (in this case probably you're library database) from whatever you're viewing the cubes with. Personally I've never played with this part but I know people who have - they've allowed people to adjust their reports' targets through the cube rather than build them an interface into the database.
Does any of that make sense? Drawback to working late is that minds gradually fade - you should see some of the SQL I just tried to run
Mike
- Thread starter
- #4
Hi,
First of all, I want to thank you. I posted my first message a month ago, when I didn't know much about OLAP. I mean I have read about it, but the things were't clear to me! I didn't know how my project should look like. I wanted to build the cube, but the essential part was that I wanted to query the cube through http, using xml. But I read that the tool for cube using http was pivot table, so I couldn't use xml. Then I decided that my project should contain 2 parts: the one that builds the cube (OLAP part) and the second one that works with OLTP.
Thank you again and if I'll have other questions, I will ask for your help
Diana
First of all, I want to thank you. I posted my first message a month ago, when I didn't know much about OLAP. I mean I have read about it, but the things were't clear to me! I didn't know how my project should look like. I wanted to build the cube, but the essential part was that I wanted to query the cube through http, using xml. But I read that the tool for cube using http was pivot table, so I couldn't use xml. Then I decided that my project should contain 2 parts: the one that builds the cube (OLAP part) and the second one that works with OLTP.
Thank you again and if I'll have other questions, I will ask for your help
Diana
- Status