Running Multiple source systems - skewing dashboard drilldown


(sivaraman gangadharan) #1

Hello ,
we have created DQD dashboard with multiple source systems , when i run one of the process involving only one source system it takes away all the drill down off other system . I do see it is both pointing to same Oracle database for these two systems and would like to know if there is way to run these two processes to run in different frequency ( daily vs monthly) and still drill down works fine .

we would like to know if it can be achieved with one datamart repository for multiple sourcesystems . It will keep one dashboard for organization level whihc an define data quality for multiple entities .

please let us know if you need more details than screen shot attached.

under bin folder :
run_processing_edwpf_patient_financials.bat … runs separately then it skews other run_processing_mrprod_casemix systems drilldown .


(Petya Minovska) #2

Hi Siva,

Usually, this should not happen. Drill down is overwritten at each run.
In order to investigate the reason for this behavior, we would recommend you to open a ticket in Helpdesk and describe the issue in more detail.

Regards,
Petya


(sivaraman gangadharan) #3

Petya ,
yes drilldown is overwritten at each run . due to this i cant run one of the 2 source systems separately which overwrites the data when process ran for both the systems .

Is there a workaround for this where DQD data mart will be updated versus completely overwritten as of now .

Thanks
Siva


(Petya Minovska) #4

Hi Siva,

As we mentioned this problem should not appear. If you are using an individual script to do the reprocessing, the Drill-through gets overwritten but the data from the different systems will not be deleted.
We would like to ask you to make sure that you are processing each system individually?
In addition, we would like to advise you to open a ticket for further investigation and share with us screenshots from ‘select count(*) from DRILLTROUGH_TABLE’ from all the Drill-through tables (tables starting with ‘DT_’) from the DQD Datamart (BEFORE and AFTER doing the system processing). You can find the Ataccama Helpdesk Manual on this link.

Regards,
Petya


(sivaraman gangadharan) #5

Petya ,
Thanks for the details . I ran one system and was able to drill down the other system as well . It seems fine , i am attaching the drill through table report as you requested .
please explain me about tables with DT_D & DT_R , what do these tables stand for .
Also . i have some 600 records in outliers which should appear in drill through table but am not able to find that number in the count attached .

DrillThrough_Tables_Analysis.xlsx (10.6 KB)

Thanks
Siva


(Petya Minovska) #6

Siva,

There are two tables per model configured in the DQD project, DT_D and DT_R table. The naming convention is DT_R_X and DT_D_X, where X is a generated random number. The DT_R tables contain information about the data quality checks and aggregation of each drill-through. Then there are the DT_D tables containing the details of the records (the actual data).
Please, note that the maximum number of records to store is 1000. The actual number of records to store is configured in the DQD project. You can check the number of records by navigating to DQD_PROJECT -> Preferences -> Output Record Sets -> Noncomplying Record Sample Size. See the screenshot below.

Regards,
Petya