Method for the continuous processing of two-level data on a system with a plurality of nodes

A method for continuous processing of two-level data on a system with a plurality of nodes for processing the data, the method comprising:

• a) determining a system state representing at least one of actual or possible performance capabilities of the system,
• b) determining already processed data on the nodes,
• c) splitting and assigning high-level input data for processing with lower level data on one or more of the plurality of nodes according to the determined system state, processing requirements of the data and already processed data in a form of at least of lower level data on the nodes such that in case of the already processed data,
• c1) data to be processed is compared with the already processed data, and
• c2) input data is split and assigned to the plurality of nodes such that an amount of data to be exchanged for processing the input data on the respective nodes is minimized,
• d) distributing the split input data to the one or more of the plurality of nodes for processing with the lower level data, and
• e) processing the distributed data on the one or more of the plurality of nodes.