Download presentation
Presentation is loading. Please wait.
1
Dataflow Machines CMPE 511
İsmail KAHRAMAN 2/12/2004
2
Presentation Outline What is data flow technique?
History of Data Flow Machines Data Flow Models Types of Data Flow Systems Macrodataflow Conclusion
3
Mechanism on Architectures
Control flow mechanism-von Neumann Data driven or dataflow mechanism Demand flow mechanism Pattern flow mechanism
4
Data Flow Technique Alternative computer design to stored program systems. The Difference Traditional Methods An instruction executed when the previous instruction in a defined sequence has been executed In this model, It is executed when the operands required become available.
5
History developed in 1960s by Karp and Miller as graphic means of representing computations. In the early 1970s Denis and others began to develop computer architectures based upon the dataflow computational model.
6
Data Flow Models It uses a directed graph called a data dependence graph or dataflow graph arcs Add [ ] + nodes [ ] Each node of a data graph can be represented as an activity template The general flow of the model is top to down
7
Terminology Data, data dependency graph Arcs : flow of the data
Nodes : represent operations Token : represent data and operations Firing: when data is available, operation is started, then result is sent to destination. Handshaking: Result is consumed
8
Data flow example
9
Switch and Merge to build condional and loop operations
10
Loop graph
11
Static dataflow Recursion is not implemented
Each node can fired only as once instance Firing rules
12
Firing rules on static Nodes fire Acknowledgement
when all input tokens available output tokens have been consumed input token are then removed output tokens are generated Acknowledgement destination sends back signal to source PE when token consumed
13
OPCODE OPERAND DESTINATION
Token format in static PE receives operation packes of the form OPCODE OPERAND DESTINATION PE sends packets of the form VALUE DESTINATION
14
static dataflow diagram
15
Dynamic data flow Firing rule
A node fires all input tokens with the same tag appear More than one token is allowed on each arc Previous output need not be consumed before the node fire again
16
Dynamic data flow architecture
17
Disadvantages of dynamic dataflow
Overhead involved in matching tokens is heavy Resourse allocation is a complicated process The long instunction cycle
18
Other types of dataflow mechanism
Pure dataflow Macro dataflow Hybrid dataflow
19
Pure Dataflow machines
20
Hybrid data flow
21
MACRO DATA FLOW
22
Conclusion Dataflow has had a fairly long development time, with a few groups studying the technique without it becoming widespread in commercial use. Compitability with existing system inhibit the introduction of a radically different computer system requiring a different style of programming and different programming languages.
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.