I could think of two ways
- use a cascade of joins in KNIME
- use a SQL connection. You might try SQLite
Work with SQLite database - try the local BigData implementation of KNIME (https://www.knime.com/whats-new-in-knime-36#create-local-big-data-env) to use if it could help you distribute the workload. Also a good exercise if you once have to move your data to a real big data environment (that gives you extra power)
Of course it depends on how large your data is. In the end KNIME might need more power. You can read about that here: https://www.knime.com/blog/optimizing-knime-workflows-for-performance
If you could upload an example we could make a test.
If you have previous experience with SAS you might profit from this booklet:
https://www.knime.com/knimepress/from-sas-to-knime