Can Data Warehouse be used as source for operational system

What are the pros and cons of using Data Warehouse (DWH) as source for an operational system?

Background:-
I work for a company who have now opted to use DWH to populate some of the small operational system's data base, The data in the DWH is populated from 5-6 huge source systems. Now that the data in DWH is integrated and has lot of derived data, Our company has begun this trend of using DWH as source data for some of the new applications that is developed. i.e using dblinks, links are created to DWH database and data in operational database is refreshed on regular basis, I somehow don't like this idea but don't have concrete points to highlight
Can anyone please let me know if this is good approach? if or if not why please?

Absolutely a good approach. How do you think Amazon pitches products that might interest you? The data warehouse of course. Teradata lumps this under the name "active data warehousing". I'm not sure using DB Links is the best implementation, but leveraging data warehouse analytics in operational applications is a great idea.

Hi,
Thanks for replying, I am sorry if I have not given much information in the first post, I would have been happy about the approach if DWH was used for analytical view point, but here they are using just as a source data

Example - We have a table in source system, this is a code table, which has around 100,000 rows with codes and its descriptions. while loading into the DWH we have made this code table into 3 columns, codes, descriptions and a derived column group.

Now, because the DWH has the group column, people want to read data from DWH and load it into another operational database instead of reading the data from the actual source system.

My view is DWH should purely be used for Analysis and Reporting purpose and should not be used a source for an operational system.

please provide more pros and cons of this approach, and will this approach create issues in the future? has anyone faced any issues when this kind of approach was used?

Data warehouses have evolved and are becoming more integrated into day-to-day operational systems. It is not a bad thing as it increases the value of the data warehouse to the organization.

I have worked on two such large scale BI projects that are aiming at that direction and they share a common main reason, people give up the hope on the old OLTP systems after having invested tens million dollars trying to patch up on the old system by outdated technologies and ending up with a mixed monster that failed to meet neither OLTP nor OLAP requirements effectively.

If DWH can deliver operational requirements, It means OLTP and OLAP systems coexist in a BI system with different modelling techniques serving for different purposes. One of the main tasks in operational requirements is data entry operation that heavily relies on normalised ER structure and RDMS constraints to ensure data integrity. Dimensional data store is to address issues in storing and querying current and historical data more efficiently. So we still need two distinctive areas in the BI system to full-fill both OLTP and OLAP requirements.

That code table idea sounds like mixing up many entities into a single table. I would break it up into more meaningfull tables, at least in dimensional store. Here is a post about this very topic: http://forum.kimballgroup.com/dimensional-modeling-and-data-architecture-f6/dimentity-containing-persons-and-companies-t772.htm

If we are talking about reporting or retrieving operational data from a BI system, we are touching on real or near real time data warehousing. One of the projects that I have been involved in required kicking off ETL process every 20 minutes between daily loads and that’s regarded as near real time DWH. Even for the real time DWH, you normaly do it by drilling across different data stores, through aggregate navigation, to provide information at different level of granularities.