SPI - a system model for heterogeneously specified embedded systems

Ziegenbein, Dirk; Richter, Kai; Ernst, Rolf; Thiele, Lothar; Teich, Jürgen

doi:10.1109/tvlsi.2002.807767

Cited by 35 publications

(17 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MoCs permit the use of efficient domain-specific optimization methods [19]. The advantages have been shown by many examples, e.g., for real time reactive systems [2] and in the signal processing domain [4,3].…”

Section: Related Workmentioning

confidence: 99%

Classification of General Data Flow Actors into Known Models of Computation

Zebelein

Falk

Haubelt

et al. 2008

2008 6th ACM/IEEE International Conference on Formal Methods and Models for Co-Design

View full text Add to dashboard Cite

Applications in the signal processing domain are often modeled by data flow graphs which contain both dynamic and static data flow actors due to heterogeneous complexity requirements. Thus, the adopted notation to model the actors must be expressive enough to accommodate dynamic data flow actors. On the other hand, treating static data flow actors like dynamic ones hinders design tools in applying domain-specific optimization methods to static parts of the model, e.g., static scheduling. In this paper, we present a general notation and a methodology to classify an actor expressed by means of this notation into the synchronous and cyclo-static data flow models of computation. This enables the use of a unified descriptive language to express the behavior of actors while still retaining the advantage to apply domain-specific optimization methods to parts of the system. In experiments we could improve both latency and throughput of a general data flow graph application using our proposed automatic classification in combination with a static single-processor scheduling approach by 57%.

show abstract

Section: Related Workmentioning

confidence: 99%

Classification of General Data Flow Actors into Known Models of Computation

Zebelein

Falk

Haubelt

et al. 2008

2008 6th ACM/IEEE International Conference on Formal Methods and Models for Co-Design

View full text Add to dashboard Cite

show abstract

“…A completely different class of approaches is based on an abstract representation of the multiprocessor system, in terms of queue networks or Petri nets, where processors are modeled as requestors, and buses and memories as queues or places [Tsuei and Vernon 1992;Lahiri 2001;Ziegenbein et al 2002;Richter et al 2003]. These approaches provide analytical performance models, which are, in principle, applicable at a very high-level of abstraction, provided that the suitable parameters can be extracted from the inspection of the application.…”

Section: Related Workmentioning

confidence: 99%

Power macromodeling of MPSoC message passing primitives

Loghi

Benini

Poncino

2007

ACM Trans. Embed. Comput. Syst.

View full text Add to dashboard Cite

Estimating the energy consumption of software in multiprocessor systems-on-chip (MPSoCs) is crucial for enabling quick evaluations of both software and hardware optimizations. However, high-level estimations should be applicable at software level, possibly constructing effective power models depending on parameters that can be extracted directly from the application characteristics. We propose a methodology for accurate analysis of power consumption of message-passing primitives in a MPSoC, and, in particular, an energy model which, in spite of its simplicity, allows to model the traffic-dependent nature of energy consumption through the use of a single, abstract parameter, namely, the size of the message exchanged.

show abstract

“…The SPI workbench [16] evaluates possible schedules for heterogeneous systems. SPI implements a non-executable (non-functional) model.…”

Section: 1: Related Workmentioning

confidence: 99%

Context-aware process networks

Dijk

Sips

Deprettere

Proceedings IEEE International Conference on Application-Specific Systems, Architectures, and Processors. ASAP 2003

View full text Add to dashboard Cite

In industry, embedded systems for stream-based 1: IntroductionEmbedded systems for stream-based processing (video, audio) are often found in consumer electronics applications. They need to operate flawlessly and hence much effort is spent in the design and verification of these applications. A popular model of computation for these applications is the Kahn process network model (KPN). The Kahn model is compositional and allows asynchronous coupling of the processing nodes in a network. An important advantage of Kahn networks is that verification of the concurrency properties of a network can be done without knowing the internal processing details of the nodes, provided that the port scans of a node are done in prescribed way.Apart from handling streams, embedded systems need to be able to handle events. A simple example of an event is a user pushing a button. Incorporating events in Kahn networks is awkward, often events are turned into a stream, e.g., a coherent stream of "on" and "off" events.However, modern embedded systems require more complex forms of handling events. Events, typically, represent changes in the context of the embedded system. The button example represents a discrete change in the mode of the operation of the embedded system, but changes of context can also be more gradual. An example of a gradual change is an increase or decrease in transmission rate of a system due to external causes (e.g. a wireless link).Proper system integration in embedded systems relies on the availability of determinate -and thus compositional -components in order to optimise functionality, performance, and resource utilisation. Unless one effectively can capture (isolate) their context dependent parts, any composition of indeterminate components yields a non-deterministic system. Although examples of working non-deterministic systems are ubiquitous, the current practice is not satisfactory.Having a proper Model of Computation (MoC) greatly improves the effectiveness of the system integration phase in developing context-aware embedded systems. Basically there exist two approaches. One either develops a MoC based on discrete-event processing or a MoC based on

show abstract

SPI - a system model for heterogeneously specified embedded systems

Cited by 35 publications

References 28 publications

Classification of General Data Flow Actors into Known Models of Computation

Classification of General Data Flow Actors into Known Models of Computation

Power macromodeling of MPSoC message passing primitives

Context-aware process networks

Contact Info

Product

Resources

About