Hi @Shubh
I believe your question is very similar or perhaps the same as in the threads below:
or may be this other one:
I made myself often the same question too, specially with complex workflows with long connections.
While waiting for a more formal solution for this, I use the following trick for the time being:
- A -Table Writer- node which writes the table in a local directory of the workflow. I also connect this node with the rest of the workflow using a variable connection at the point I want it to get it executed (synchronized).
- At the other end of the workflow a -Table Reader- node which reads the table and brings it back to the workflow. This node is also connected to the end of the workflow by a variable connection so that it is only executed at the right time.
In this way, the data is conserved from a point (1) to a point (2) of the workflow without a long connection, while keeping it synchronized with the rest of the workflow.
The following snapshot shows schematically how to implement it:
May be this could help in your case.
Best
Ael