DATA FLOW COMPUTERS • Data flow computers execute the instructions as the data becomes available • Data Flow architectures are highly asynchronous • In the data flow architecture there is no need to store intermediate or final results, because they are passed as tokens among instructions.
(cont. ) • The program sequencing depends on the data availability • The information appears as operation packets and data tokens • Operation packet = opcode + operands + destination • Data tokens = data(result) +destination
(cont. ) • Data flow computers have packet communication architecture • Dataflow computers have distributed multiprocessor organization
Classification based on data handling • Static Model • Dynamic Model
Static Model – Tokens are not labeled – Uses control tokens for appropriate timing and control
Dynamic Model • Uses tagged tokens
Common to both models • Both models have a pipeline ring structure with the following components: – The memory – The processors – The routing network – The input/output unit
Technical Problems to be solved • Develop eficient programming languages • Development of the decomposition of programs into modules and module assignment techniques. • Better control and support of intercommunication packet switched networks
(cont. ) • Development of intelligent data driven mechanisms • Efficient handling of complex data structures • Better memory organization for support of data flow • More accurate performance evaluation