on 12-24-2009 10:53 AM
What is the diffrence between delete/create index from cube step present in process chain and delete/create BIA index of cube?
Hi Gaurav,
DB Index
Database index is created on the database tables to improve read and write to the cubes.When the indexes are deleted for any cube while loading data to the cube it does not need to search the index for writing in the cube.Hence the loading performance is improved.Hence when we say we are deleting the DB index from the cube actually we are deleting these database indexes similarly while creating the indexes for the cube , we are creating these database indexes which inturn will improve the read performance.
The frequency of Deletion and Generation of Indexes depends on the volume of data the Info cube holds. If the cube has less data, the Indexes can be dropped before every load and Generated after the load is complete. Otherwise you can schedule a weekly run for the Deletion and Generation of Indexes in case of large volume of data.
BWA Index
You can create only one SAP NetWeaver BI accelerator index for each InfoCube. This BI accelerator index contains all the data from the InfoCube. In contrast to the procedure with aggregates, it is not necessary to make specific selections and restrictions for the definition of the BI accelerator index.
BWA indexes are created for the cube to solely improve the query performance and has no impact on the dataloads.
And you need to rollup the data after new data is loaded. I am assuming in your case , you might be doing some deletions hence deleting and recreating the BWA index.For more details on this you can refer a very good blog at the link below.
/people/sara.hollister/blog/2008/07/20/bia-blog-series-part-iv-bia-monitoring-and-maintenance
Hope this answered you.
-Vikram
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hello Vikram,
I have a Business related issue, where the Business reports that the PRODHIER LEVEL6 value that is shown in the report is not correct.
Instead of showing the whole length of data in the field, it is showing only its first two characters.
Ex. PRODHIER Value - F42XXXXXXX
And PRODHIER6 should display F42XXXX, but at report level and at cube level after doing DB aggregation the value that is displayed is F4.
I tried dropping BIA Indexes and rebuilding but the issue is same even then.
I have thought of an alternate solution but haven't tried it yet. The solution is to drop the data at the cube level and then reload it from the base ODS and after rollup, to check once again. Please advise if this is feasible or if you have an alternate solution, please advise.
Is PRODH6 a navigation attribute in the cube or is it a direct field in the cube? If it is navigation attribute in the cube, what is it showing in data display of cube data in RSA1? What it shows here in RSA1 can't be different to what exists is master data in case of a navigation attribute.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
83 | |
24 | |
12 | |
9 | |
7 | |
6 | |
5 | |
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.