on 11-16-2017 10:06 AM
Hi all,
Is it possible to create a datasource delta enabled on a HANA calculation view containing data coming from SDI data replication? We want to use this datasource to load data in our BW 4/HANA system and we want to avoid the multiple loads of the same records.
An idea could be something like a remote drop table call from BW at the end of our process chain to empty the replicated table on HANA. But we don't know if this way is possible.
Thank you.
Thanks Maksim
But how are you able to process the data package by package
Normally entire calculation view gets generated and then DTP process it package wise
Sometimes we are getting memory dump when data being prepared by calculation view
Thanks
Deb Deep ray
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi, you could create datasources in BW/4 in ODP_HANA source system.
t could be used for loading data from tables,views, HANA Models (Calculation Views included). Delta is supported through ODQ at SAP BW/4 side. Configure a delta when you define datasource in BW/4, you have to base on field (Data, Timestamp or Numeric Pointer).
Br, Maksim
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
We can create datasource delta enabled both in HANA_LOCAL and in ODP_HANA source systems, but what's the difference?
Furthermore what's the difference between:
-Pure additive delta
-Overwrite delta
If we use a delta field type "date" and we run more time on the same day the load what's the result?
We would like to execute more time on a day the load and extract all the time the records related to that posting day, using the overwrite method to be sure that records which have been already extracted don't get duplicated amount/quantity.
Thank you.
From my point of view, there is a simple logic.
- If you need DB table or DB view, created in SQL Editor for example, then use HANA_* system types (e.i. HANA_LOCAL).
- For loading HANA Information Models, previously you asked about HANA Calculation views, use ODP_HANA system.
About "date" field and loading more than 1 time at a day. Try to find more granular timestamp and use it as delta field. If there is not such option, configure safety interval (lower = 1 day). This will result to duplicate of data records, load them with overwrite option to aDSO. In this case additive delta is not possible.
BR, Maksim
Hi guys,
First of all, I want to tank you for your answers. Below the description of our scenario.
We solved this issue using the replication task with the feature load behavior = REPLICATE WITH LOGICAL DELETE.
The result of this setting was the creation in our replicated table of a timestamp field, filled with the timing of the data load.
Since our replication task is real time, we get populated this field with groups of records with same tmstp.
We imported this field in our calc. views until the semantic layer and use it in our datasources (that are based on these CV) in BW as delta short timestamp with the proper UTC+n fuse. We tested this solution and it works properly, allowing us to receive the fresh data every time we run the process chain on BW side (actually every hour).
Thank you again for support.
By the way we did not undestand the difference between Hana Local and ODP_HANA 😞
User | Count |
---|---|
83 | |
10 | |
10 | |
9 | |
7 | |
6 | |
5 | |
5 | |
4 | |
3 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.