I think you are over thinking on a potential issue. The size of tables as you said will be rarely over one million records which does not qualify it to be considered a big table. In my application we have tables which are much bigger than that but still work pretty well over time. Also, as you said that data is mainly categorized on a daily basis so, theoretically it makes sense to partition tables on dates.  It should improve the performance to some extent. But it can also lead to good number of partitions which can backfire too from ... Read More

Projection Node is basically a graphical Calculation view. SAP HANA does not allow usage of regular expression in a graphical calculation view for filtering out data.So if you still need to use graphical calculation view (Projection Node), then you need to filter out the data either by using a table function or equivalent.

I don’t think there exists any way where you can define the column type of a column within a table type as an array. Table type refers to a row structure with predefined number of columns. So you might need to rethink on how to change the incoming data from .NET application to be a usable one for your SAP HANA project.

I am not sure whether someone can get you exact answer, SAP is not open source and implementation details and bits are not known to many except the creators and few others. If I go with the standard definition, for storage you will require the size of length of the data plus number of bytes for header information. So as per your example it will require 9 + 2 (for header) = 11 overall for storage.

I am not sure that there exists something in SAP HR in pure form which supports your use case. But you can handle the scenario as mentioned by you by making some changes here and there.There exists an infotype ‘Contracts’ with SAP HR which stores the type of employment of the employee like permanent or contract. You can take care of this type by flipping the status of the employee type by some third party service.  Once the employee resigns then rehire him as a contractor but create a dummy pay roll so that it does not gets picked up ... Read More

As you said you are using the hot folder concept, you need to ensure that the proper threads are being used at every moment for any operation.By default, the number of threads used for import is one. To be more technical, impex.import.workers thread is one if not set explicitly. Since in your case the file is big, make sure that you exploit all the available cores to the fullest. It is highly recommended that maximum threads at any point of time should not be more than twice the number of cores. For optimal working, you may need to work out ... Read More

Basically Infostore refers to the metadata about Objects and in BO terms – infoobject. An infoobject is basically an entity in SAP business Object universe. It can be a report or a user or an event as well.For E.g.: if we are speaking about an employee so an employee has a metadata like Name, Department, Designation, Salary and others. Infostore lists out this metadata when the context in Employee. So when you execute a GET request for getting infostore it returned you the metadata for an info object.

The logic is pretty simple. You can define a local variable which is used for storing the total sum of all the groups. When you are performing iteration for each group, you can add the sum at the group level to the local variable.I am not suggesting code as it is very basic and straightforward.

Yes, there exists an equivalent. You can go for using SAP HANA Smart Data Integration also known as SDI. It is a standard offering of SAP for data integration.  You don’t require any additional infrastructure for this, like a separate server or other. There are lots of blogs and online help which can help you to start on SDI.You can also check out for Data services but that is too complex and may be excessive for the mentioned usage.

Yes, the PBO-PAI triggers commit so your requirement will fail. You need to make changes to your current implementation to avoid the auto-commit.