Best Of

Hello,

Extract cube works in version 11 the same way it was working in version 10: it extract the cubes in a text file sorting the file columns in the most efficient way.

The most efficient way in Board 10 was depending on 2 factors: MAX item number and sparsity.

Board 11 does not have Max item number and sparsity so the column order is different.

Example if you have 3 entities A,B and C in the structure of your cube in Board 10 the headers of the extracted file will look like

A   B   C value

In Board 11 this might be unchanged or it might vary, for example it can be

B   C   A value

Because by default the extract cube uses the most efficient extract

In the procedure step extract cube of board 11 we introduced the possibility to select the columns and their order, so in case you have such a discontinuity you can select the column order from the procedure step.

For example i am extracting the cube gross sales by Month, Customer and Product

I can also decide to extract in a different order, for example Customer, Month and Product

Or i can do something that was not possible in previous versions like extracting it by Year, Month and Customer

I know that it might take some time to review all these procedure steps, but we created this functionality to make this transition as smooth as possible.

This functionality also solves another issue: in Board 10, in case of a change in Max item number the extract might vary, using this functionality keeps the format of the file no matter what, so from now on you should not face this problem anymore.

Hope this is clear,

Antonio

Hello Daniele,

Join algorithm is only supported in InRam Mode.

If you are running HYbrid you need to set the three involved cubes as InRam and restart the service, or run the serice in full InRam mode.

Jonathan Baetens The adulaparams caches are only used for dataentry.

Ciao

Antonio

Hello Paul,

The copy paste of layout and objects is only for me and VIPs, in fact I have it (see the screenshot below)

You Need to have 5,000 community points to unlock the feature

Of course I am kidding, it's true that you do not have the copy paste in your 10.5, but it's also true that I do have it on the upcoming version.

the reason is that the feature is arriving: board web can be updated now independently from Board Engine, so the upgrade of new releases will be very fast (it's not a winclient development luckily).

You will soon have a new release that can be installed in no time and copy/paste will be there with more and more features that will be available.

Stay tuned!

Hello Sebastian,

There are some options to calculate the month to date.

Usually, when I need to calculate the month to date on a day cube, I create a copy of the cube by month, then I use the year to date on both cubes and use an offset -1 on the month cube, the difference between the two cubes will be the month to date. This logic fails in january (the offset -1 is the reason) so we also need a cube to identify if we are doing this calculation in a day in January or not.

This logic works also in combination with the previous year function, so if you need to make a comparison with previous year you can do that.

The Calendar enhancements are not designed for that, but there is a way to use them to get the month to date. I do not suggest it, but if you are curious just let me know.

I shared an Inspiration for the first solution, you might take a look

An example of Month to date calculation

Hi,

When you use an algorithm in a dataview to calculate the result of a function, data is aggregated first, then the function is applied.

When you use the dataflow, the function will be performed at the maximum detail level (cell level - therefore per every element of every dimension of the cube), and only after having calculated the result of the formula per every cell, data is aggregated at the level you are displaying it into the dataview.

At the maximum granular level the dataflow and the algorithm will show the same result, but as you aggregate values and dimensions (by default BOARD aggregates data by sum), you will see the sum of the product [dataflow] instead of the product of the sum [algorithm].

Hope this helped,

Regards,

Antonio Speca,

Product specialist

Hi,
My guess is that you have created a Database Security Profile, but you probably forgot to associate it with your user (thus, locking yourself outside of the database).

In the following BOARD Manual section, you can find how more info on Security concepts and Database security profiles
Database security concepts 
when you create one (and most importantly when you create the user profile - administrator), you should assign it to the a security profile and a to user
How to create a user profile 

Please let me know if this helped you fixing the situation.
Regards,
Antonio

Hello,

Every sum algorithm is based on binary numbers, that's why you might incur in situations where a sum has those little discrepancies. Try to put a ROUND function in your dataflow when calculating C block to avoid this issue.

e.g.

C=ROUND(A+B,2) > this should round your numbers to the second decimal digit and solve the issue.

Considering the situation in a broader perspective, we can say that a number with a finite number of decimal digits in decimal base, may have an infinite number of decimal digits in its binary expression.
Moreover, the floating point (applied in BOARD) gives extra precision to the result when it's zero because it uses less digits for the non-decimal part. Since almost every computing engine (and this includes BOARD) uses binary expressions of numbers and finite number of decimal digits, the above consideration should be taken into account when handling very large numbers in each cell (trillions/billions with high number of decimals).

Hope this helped,

Antonio Speca

Product specialist