Can't we use the same source objects if it is already being used in different replication flow?

[Rupashree]

If not then why ?

[NicolasMartinez]

That's also a problem for us.

Fortunately it should be solved soon, have a look at the following roadmap item:

Allowing the reuse of source data sets in multiple replication flows (Planned for Q2 2025) 

I hope that it will also allow to load the same object from the same source system in two different replication flows.

[Martin_Kuma]

Why do you need it? Basically you double the data. This is contrary to the single read principle. Anyway you can bypass by creating second connection (from second space) and then share the table.

[NicolasMartinez]

We need to load ACDOCA with the relevant CDS View into our Datasphere. The initial load is too big when we try to do it for the last four years, and it always fails.

Initially, we loaded each historic year as a full load and do a delta only for the current year. However, the full loads / initials take around eight hours.

We then realised that the customer has changes in ACDOCA for previous years too, which is why it isn't enough to only have the current year as a delta.

However, regularly loading every year as a full load with an 8-hour runtime is also not a viable solution.

That's why we plan to use several replication flows, each for one year, and use the delta for historic years as well.

[Rupashree]

Great, Thanks

[Martin_Kuma]

Hi Nicolas,

I see. I would split the load even more then. Not only years, but comp. code for example as well. For the past years at least. 

Definitelly a problem. But the feature has just been installed / released 🙂

Cheers, 

Martin 

[NicolasMartinez]

We already tried it and it works 🙂