I´m working on an 8 kernel-linux system with the latest knime version.
I want to use the molecule fragmenter node for a set of 60 000 compounds !
Even if I split the fragmentation process on all CPUs (7500 Molecules per CPU) by prefixing some row splitter nodes before and run the 8 fragmentation nodes in parallel, the process is even not finished after 15 hours.
I also think that by enlarging (or reducing) the size of the fragments the run time also won´t abbreviate significantly...
Does anybody have an idea to overcome this run time problem or is this just a characteristic of the fragmenter node?
Thanks in advance !