DataFlow in KNIME provides many nodes that are built using DataFlow. These nodes can be available in the DataFlow category in the node repository.

DataFlow in KNIME also includes many native KNIME nodes. Native KNIME nodes generally run in the DataFlow executor and will not run by default in a cluster. To enable the KNIME nodes to run in a cluster, you must change the default settings for plugins that are supported in the cluster. In addition, add the -Dknime.swt.disableEventLoopWorkaround=true argument in the -vmargs section in the local client. This argument must be included to the JVM arguments for the master and workers of cluster using the job.master.jvm.arguments and job.worker.jvm.arguments settings.

For more information about defining the plugins that must be automatically deployed while executing in the cluster, see Cluster Plugin Preferences.