A scenario-aware data flow model for combined long-run average and worst-case performance analysis

Theelen, B.D.; Geilen, M.C.W.; Basten, Twan; Voeten, J.P.M.; Gheorghita, S.V.; Stuijk, Sander

doi:10.1109/memcod.2006.1695924

“…In specific subclasses of the RPN model, this synchronization may be realized without much overhead. There are very regular subclasses of RPN, such as the SDF-based models of Scenario Aware Data Flow (SADF) [48] or Heterochronous Data Flow (HDF) [21]. SDF graphs execute periodic behavior, often called iterations.…”

Section: Operational Semantics Of Rpnmentioning

confidence: 99%

“…In *charts [21], which can be seen as an RPN instance, specification is done by direct refinement of FSM states by dataflow graphs. In SADF [48] actors or processes read control tokens sent to them by the controlling FSM and adapt or suspend their behavior accordingly.…”

Section: Dmentioning

confidence: 99%

“…The Markov model of the transitions between frame types can be used to study the expected performance of the given system [48]. Alternatively, a worst-case analysis can be done to establish a guaranteed performance using techniques such as introduced in [42] and [18].…”

Section: Analyzable Models Embedded In Rpnmentioning

confidence: 99%

“…Many of the combinations of data-flow and reactive behavior are based on a combination of the Synchronous Data Flow model together with some form of reactive behavior [21,29,[47][48][49]. The use of an analyzable model such as SDF is natural because it allows for a predictable and determinate combination.…”

Section: Bibliographymentioning

confidence: 99%

“…The goal is to be also able to incorporate more analyzable, less expressive models in the same way SDF fits within the KPN model, but still combining data-flow and control oriented behavior. For instance, a combination of SDF with finite state machines such as HDF or SADF [21], [48], [47], which yields analyzable models, would fit within the framework. This is illustrated in Figure 9.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Kahn Process Networks and a Reactive Extension

Geilen

¹

,

Basten

²

2013

Handbook of Signal Processing Systems

View full text Add to dashboard Cite

Kahn and MacQueen have introduced a generic class of determinate asynchronous data-flow applications, called Kahn Process Networks (KPNs) with an elegant mathematical model and semantics in terms of Scott-continuous functions on data streams together with an implementation model of independent asynchronous sequential programs communicating through FIFO buffers with blocking read and non-blocking write operations. The two are related by the Kahn Principle which states that a realization according to the implementation model behaves as predicted by the mathematical function. Additional steps are required to arrive at an actual implementation of a KPN to take care of scheduling of independent processes on a single processor and to manage communication buffers. Because of the expressiveness of the KPN model, buffer sizes and schedules cannot be determined at design time in general and require dynamic run-time system support. Constraints are discussed that need to be placed on such system support so as to maintain the Kahn Principle. We then discuss a possible extension of the KPN model to include the possibility for sporadic, reactive behavior which is not possible in the standard model. The extended model is called Reactive Process Networks. We introduce its semantics, look at analyzability and at more constrained data-flow models combined with reactive behavior.

show abstract

Kahn Process Networks and a Reactive Extension

Geilen

¹

,

Basten

²

2010

Handbook of Signal Processing Systems

View full text Add to dashboard Cite

Kahn and MacQueen have introduced a generic class of determinate asynchronous data-flow applications, called Kahn Process Networks (KPNs) with an elegant mathematical model and semantics in terms of Scott-continuous functions on data streams together with an implementation model of independent asynchronous sequential programs communicating through FIFO buffers with blocking read and non-blocking write operations. The two are related by the Kahn Principle which states that a realization according to the implementation model behaves as predicted by the mathematical function. Additional steps are required to arrive at an actual implementation of a KPN to take care of scheduling of independent processes on a single processor and to manage communication buffers. Because of the expressiveness of the KPN model, buffer sizes and schedules cannot be determined at design time in general and require dynamic run-time system support. Constraints are discussed that need to be placed on such system support so as to maintain the Kahn Principle. We then discuss a possible extension of the KPN model to include the possibility for sporadic, reactive behavior which is not possible in the standard model. The extended model is called Reactive Process Networks. We introduce its semantics, look at analyzability and at more constrained data-flow models combined with reactive behavior.

show abstract