BOARD 11.1 - Dataflow not working

Dear all,

 

we are migrating a database from 10.3 to 11.1 but dataflow do not work. We start to check this dataflow


Net Sales (A) + Credit Notes (B) = Sales (C)

 

A and B has these entities: Day, Document No, Order, Item

 

C has these entities: Day, Customer, Order, Item

 

Document No and Customer are into a relationship. All relationship are checked and normalised. Consider that these dataflow "opens" new tuples of data because A has day1,item1,doc1 and B could have day1,item2,doc2.

 

We tried to perform these procedure:

 

Case 1 - NOT WORKING

 

A+B = C changing both calculus dominion and sparsity properties

 

image

 

 

Case 2 - NOT WORKING

 

We tried to define temporary cube A' and B' with same entities of C so the procedure became

 

A' = A

 

B' = B

 

C = A' + B'

 

Case 3 - NOT WORKING

 

We tried to open sparsity using

 

C = A

 

C = B

 

A + B = C

 

in these case we also experienced that BOARD performed only C = a, ignoring following steps.

 

Case 4 - NOT WORKING

 

We tried to open sparsity using

 

C = A

 

C = C + B

 

In all of the previous cases we tried also to select just a customer in a limited period of time (2 months) but the procedure does not work. We try also to define the dataflow into a dataview and perform a Learning Mode to create version of cube. As not working I mean:

 

- in some cases we experienced that BoardEngine and BoardWebApiEngine stopped automatically without user did it

 

- in some cases we had to stop it because BOARD was not responding

 

- in some cases BOARD performed the dataflow but number were strange

 

Checking on log I saw that it occurred Tuples.[Intersect]+ that probably is a new algorithm, I was not able to find any references about it.

Tagged:

Answers

  • I experienced the same issues when I try to copy from cubes. For example you have a cube on day level and want to copy it to a cube on month level. My solution is to make a temp cube based on the dimensions of your source cube except you change the time from day to month. In the first dataflow you fill the temp cube and in the second dataflow you copy from the temp cube to your target cube.

     

    best regards,

     

    RJ