Hi,

with Talend mostly all is possible

You could define your input parameters as context variables and calculate the values you need for processing in a tJavaRow for example. Store the results in a context variable again. To get the number of rows in the table you could use a t[DB]Input "SELECT count(*) FROM XXX".
Then connect the subjob to process your data with onSubjobOk.
To handle the blocks there are several ways. One would be to calculate the "commit points" dynamically (for example "100,200,300), store them in a context (string), use a tFixedFlowInput to create a flow, process them with tNormalize (you now have three rows). Next step a tFlowToIterate and it is done...

Bye
Volker

Hi Andrei,

yes it is. I'm pretty sure that there is a thread in the forum about this issue with an example. But I didn't find it :-( Someone else remember the thread and know how to search for?

The idea was to define a table with information which data should be processed (for example key ranges). Than read the data and create an iteration from the flow to execute you t[DB]Input with a dynamic SQL statement.

Bye
Volker

Hi Andrei,

to answer your question we need some more information.

But I think it will end up in creating one process for each table:
t[DB]Input --(row)--> tMap[or any complexer transformation] --(row)--> t[DB]Output

Bye
Volker

If your need is as simple as you wrote then backup and restore your database into your new host.