cancel
Showing results for 
Search instead for 
Did you mean: 

DumpLoad Task vs. Manual Processing

Former Member
0 Kudos

Hey Experts!

I apologize in advance for my ignorance, but I've only started working with BPC in the past month and I'm reviewing our current processes and looking for ways to improve things.

Currently we have an SSIS Package that loads quite a bit of data into one of our AppSets and at the end of the process, it kicks off a DumpLoad Task to Process the Cube utilizing the "Full Process" option.

After this package with the DumpLoad Task is run, we have an analyst who goes into the Admin Console, clicks "Manage Dimensions", chooses the same AppSet, clicks "Process Dimension", Chooses "Application" in the Select By: radio button, does not check "Take System Offline", does not check "Process members from member sheet", and does check "Full Process". He believes this last step is necessary to make sure the cube is completely processed.

My initial thought is that this last step is a redundant process. It seems from the documentation I've read, that the DumpLoad task should be fully processing the cube when the SSIS Package is run. And the additional step to process the application manually seems like it may be unnecessary. Can anyone confirm this? Or is there some reason why the application needs to be manually processed after loading new data? Is it possible that the DumpLoad Task is not as thorough as the manual process?

Any insight is greatly appreciated!!!

Thanks!

Accepted Solutions (1)

Accepted Solutions (1)

Former Member
0 Kudos

In my experience the DumpLoad task in and of itself does not perform the processing. But I have used the Custom BPC tasks to perform the Manual process of processing the cube, automatically in the package itself. The BPC System available task can be added after the DumpLoad task to take the system to the Not Available state. Then I've used the Admin task to perform an optimize after the load. However, you could use the Admin task to perform a process of the application if you felt that was necessary. Then the final step would be to use another BPC System available task to bring the system back to the Avialable state.

Answers (1)

Answers (1)

Former Member
0 Kudos

The full process option ensures that the SSAS cube is in synch with the data in the fact tables. It is not necessary to run the additional full process task unless changes have occurred to dimension members or data.

As a best practice, it is good design to include the tasks for taking the system offline and online when running a full process - especially if you're processing metadata changes or loading a lot of data that requires logical processing. The idea is to enforce read consistency as the data is loaded or members added/maintained and eliminate confusion from an end user perspective.

Have a good one!

Former Member
0 Kudos

Hmmm.. I now have more questions.

If the Admin Task using the "Process Application" option with a "Full" process method is the way to fully process an application using SSIS. Then what is the difference between running that and running a "Full Process" when processing a Cube?

How would I process a Cube in BPC differently from Processing an Application?

Thanks for your help, thus far....