Heterogeneous Design in Functional DIF

Abstract

Abstract. Dataflow formalisms have provided designers of digital signal processing systems with analysis and optimizations for many years. As system complexity increases, designers are relying on more types of dataflow models to describe applications while retaining these implementation benefits. The semantic range of DSP-oriented dataflow models has expanded to cover heterogeneous models and dynamic applications, but efficient design, simulation, and scheduling of such applications has not. To facilitate implementing heterogeneous applications, we utilize a new dataflow model of computation and show how actors designed in other dataflow models are directly supported by this framework, allowing system designers to immediately compose and simulate actors from different models. Using an example, we show how this approach can be applied to quickly describe and functionally simulate a heterogeneous dataflowbased application such that a designer may analyze and tune trade-offs among different models and schedules for simulation time, memory consumption, and schedule size.

Parameterized Sets of Dataflow Modes And Their Application to Implementation of Cognitive Radio Systems

“…A general scheduling approach for CFDF graphs is the so-called canonical scheduling approach discussed in [16]. In canonical scheduling, a sequential ordering L of the dataflow graph actors is constructed [16].…”

“…In canonical scheduling, a sequential ordering L of the dataflow graph actors is constructed [16]. At run-time, the scheduler iteratively traverses the list L, and upon visiting each actor A, the scheduler checks the enabling condition (availability of sufficient input data) for A, and invokes A if the enabling condition is satisfied.…”

“…CFDF applies the concept of actor "modes", where different modes can have differing dataflow behavior, and mode transitions can be data-dependent. CFDF is tailored to natural design of actors with dynamic functionality, and facilitates prototyping of dataflow applications, as well as identification of more specialized dataflow behaviors [16], such as BDF, cyclo-static dataflow (CSDF) [5] or synchronous dataflow (SDF) [13].…”

Modeling and optimization of dynamic signal processing in resource-aware sensor networks

“…The DSG representation, as introduced in [24], is formulated in terms of CFDF semantics, which is highly expressive and subsumes many other useful dataflow models as special cases [21]. However, the DSG representation can be extended or adapted to work with other modeling techniques; see [24] for further details.…”

Stream Compilation for Real-Time Embedded Multicore Systems

“…There are other projects, such as the Ptolemy [16], PeaCE [17], and DIF [18], that support multiple different dataflow models. This means that they cannot generate fully static run-time schedules, and have to provide run-time scheduler for run-time execution of dataflow actors.…”