I use MSSQL and have a knime procedure to process data into a database table. Most of the information for the table is an insert, however sometimes old data is re-processed. This is why I am using dbmerge as it manages both insert and update. Running an update of my table 4000 rows and 12 columns takes multiple hours (approx 4). Reading the DB or extracting the data on the other hand is really fast. 10 secs for approx 1 million rows. DB size is approx 1.8 million rows. the table is organised with a key field (as the concatenation of the 9 string fields), the 9 string fields which all must be unique and then two value fields as results for these keys.
I would appreciate any help.
Try to stream the node.