There are various factors that affect the run time for a process in Duco.
- your input size is large, say 50,000+ records per side
- you have many partial and unmatched results
... this probably means that your process is still not fully configured and you still need to improve your matching rules.
When a process is fully configured, the process will run much more quickly.
Setting up a process when a run takes a long time (10+ minutes) can be slow. A good way to proceed is often to work initially with a small input data set. You can set up your rules working with a small data set, say less than 10,000 records per side. Once you get a high match rate, you can run the process with the full input data set.
To obtain a smaller input data set, you can just create a cut-down version of your input files.
Alternatively you can use a filter rule to obtain the same result. For example the rule: "Currency is not equal to “CHF” " will exclude all records where the the currency is not "CHF" and therefore process only the records where the currency is "CHF".
The fuzzy match rule can have a significant effect on performance. Fuzzy matching is a computationally intense operation. As when working with a large data set, you can first test your process set up with a smaller data set, and then run the process with the full data set.