摘要 |
This device (10) for digital data processing based on a data flow processing model is suitable for the execution, in a distributed manner on multiple calculation nodes (16, 18, 20, 24, 26, 28, 30, 34, 36, 38, 44, 48), of multiple data processing operations modelled by directed graphs, where two different processing operations can include at least one common calculation node (16, 20, 26, 28, 30, 34, 36, 38). It includes means (12) for the identification, from a valued directed multi-graph made up of the union of several distinct processing graphs and divided into several valued directed sub-multi-graphs (54, 56, 58), called chunks, and whose input and output nodes are buffer memory nodes of the multi-graph, of a coordination module (16, 26, 34) for each chunk. Furthermore each identified coordination module is designed to synchronise portions of processing operations destined to be executed in the chunk with which it is associated, independently of the portions of processing operations destined to be executed in the other chunks.
|