In this topic we will discuss about the main ASDO’s and their functionality which will help to understand the basic BW project design in the next blog I am going to discuss about.

The below is very important . Common ADSO’s available in BW 7.5 on HANA, BW4/HANA 1.0 and  BW/4HANA 2.0 –

1. Data Mart ADSO : 

Memorable point : 

• Can be report on top of active data table and Inbound table either using AND or OR.
• When reading data using navigation the data will not change i.e stable.
• Data are ADDITIVE not OVERWRITE.
• All key form the Composite Key.
• Only in inbound table RequestID is the key like other ADSO.
• Change Log exists but not loaded.

2. Standard ADSO : 

Memorable point : 

• Data store at document level .
• Can be report on top of active data table only.
• When reading data using navigation the data will change if load happen i.e non stable.
• Data are always OVERWRITE.
• All key form the Composite Key at Active table .
• Only in inbound table RequestID is the key like other ADSO.
• Change Log exists and  loaded from inbound queue during activation . Delta DTP takes data from change log table.
• Full Load DTP takes data from Active table.

Standard ADSO with Snapshot support always do Reverse image if you full load the data next time .

Suppose you have 10 record in load1 and if you have snapshot support and there are 9 records in second row. 1 row will be deleted from ADSO but without snapshot support if you load the roq will be 10.

3. Staging ADSO or Write optimized DSO:

Memorable point : 

• Commonly used as PSA  at data entry layer . It act as a consistent base level data store.
• Field based modeling preferable for faster performance.
• Reporting not preferable.
• No key is Mandatory.
• Only in inbound table RequestID is the key like other ADSO.
• Change Log exists but not loaded from inbound queue.
• Full and delta Load DTP takes data from inbound table.

4. Direct update ADSO :

Memorable point : 

• Directly Active table updated.
• No need to Activation.
• Using request you can delete data.
• Using DTP or API data can be loaded.
• Supports only full load. After the data load the data is visible immediately from reporting.

5. Inventory Type ADSO : you can used standard and Data Mart ADSO as Inventory ADSO by enabling checkbox inventory enabled for inventory purpose use. This is not used frequently.

Note : ADSO can contain maximum 2 billion of record. For larger implementation do Symantec partitioning or Physical partitioning.

As per my recommendation :

1. Do tearing management of Hot, Warm and cold data based on the date range ( 1 year or two year)
2. if Tearing is not possible then go with Symantec partitioning.
3. if Symantec partitioning is not possible go with manual secondary partitioning of the ADSO at table level. because Primary level partitioning to all the table done automatically. You can only manually partition the Active table by range.

Hope this will help. In the next blog using this concept I will discuss about a standard,common  BW/BI architecture of a project a developer can adopt.