Force Consolidate Methods

[xtensibilitybro]

Hi Everyone,

Thanks in advance for the responses. I am reviewing the impacts on a force consolidate on a specific entity and noticed that my results differ when i run a force conoslidate from a cube view and a force consolidate using the calculation definition within process cube. 

I made sure that my POV was identical on the cube view versus the process cube. 

Can someon explain to me what the difference is or how there could be a difference between the two methods?

Thank you!

[TheJonG]

There should be no difference. Can you explain the differences in results that you are seeing? Perhaps it is a Formula Pass issue. Also screenshot the Calculation Definition and POV of Cube View if possible.

[TonyToniTone]

Agree with TheJonG.  Comparing apples to apples testing, the consolidation times for a force consolidation executed from a Cube View vs. Calc Definition should be negligible.  Maybe +- a few seconds.  Without more details or screenshots, I can only provide general reasons on why you may be seeing a difference.

1 .  The Calculation Definition has additional steps to perform outside of the Force Consolidation or the Force Consolidation points between the 2 are not the same

2.  One of the processes was the 1st time a consolidation was done for the month.  All subsequent consolidations after the 1st consolidation for the month are faster.  

3.  The force consolidations were executed on different Consolidation servers.  Check to see if both force consolidations were performed on the same Consolidation server.  

• Consolidation servers should have consistent hardware such as 12 logical processors (CPU's) and 64 GB of RAM for both servers.  If this is not the case, either process could be affected
• Each CPU has a microprocessor speed.  If CPU's have different speed microprocessors across the consolidation servers, either process could be affected.  
• CPU's have a CPU Benchmark which estimates how many millions of instructions that can be processed in a second (MIPS).  The higher the CPU Benchmark score, the faster the consolidation will execute on the consolidation server.  For cloud installations, you can find the CPU Benchmark score under System > Environment.  Click on the Cons Servers under Application Server Sets. Click on the Hardware tab and there will be a row called CPU Benchmark.  The CPU Benchmark scores test how much work a CPU can handle and compute on the consolidation servers.  If you see a difference of +- 2 points between servers, it probably won't be enough of a significant impact to consolidation times.  However, if the difference is more than that, you may see a bigger gap in the consolidation time differences.  A good score is around the 100 mark.  

4.  Is the Force Consolidation in a Calc Definition triggered through a Data Management Step and Sequence?  If so, the consolidation would execute on a Data Management server while the Cube View Force Consolidation would execute on a Consolidation server.  2 different servers which goes back to item 3 above about different hardware configuration.  

5.  Were both of the consolidations performed with no user activity in the system?  If not, there could have been a slight delay running 1 or both of the processes.  Other users may have been executing calculations, translations, or consolidations during the same time of your consolidations.  This situation could have put your consolidations in a temporary queueing phase waiting for CPU resources to free up.  

There is a lot to think about to answer your question accurately.  However, functionally, if you are testing in a vacuum, there should be very little to no consolidation time difference between execution through those 2 processes.