Design-Trotter: a multimedia embedded systems design space exploration tool

Moullec, Yannick Le; Diguet, Jean Philippe; Philippe, Jean-Luc

doi:10.1109/mmsp.2002.1203342

Cited by 7 publications

(9 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The work presented in this paper is part of a complete design framework called Design Trotter [1]. Design Trotter is a set of cooperative tools which aim at guiding embedded system designers early in the design flow by means of design space exploration, as summarized in Fig.…”

Section: Overview Of the Design Trotter Frameworkmentioning

confidence: 99%

“…For each function, the two first steps (presented in this paper) include the construction of the graph (HCDFG model) and the characterization by means of metric computations. Then a scheduling step, presented in [1] is performed for all the data-flow graphs and loop nests within the functions. The design space exploration is performed by means of a large set of time constraints (e.g., from the critical path to the sequential execution).…”

Section: Overview Of the Design Trotter Frameworkmentioning

confidence: 99%

“…The characterization of the functions has two objectives: (i) guide the designer in his architectural choices and (ii) guide the function estimation step in order to use the most adequate scheduling algorithms [1].…”

Section: Characterizationmentioning

confidence: 99%

“…A local access is a data-transfer to/from a data defined in the considered graph. N1 data are always global, N4 data always local, N2 and N3 data are initially local but can generate local/global swapping during the scheduling steps if the local memory size is limited [1].…”

Section: Memory Modelmentioning

confidence: 99%

See 3 more Smart Citations

Algorithmic-level Specification and Characterization of Embedded Multimedia Applications with Design Trotter

Moullec

Diguet

Amor

et al. 2006

J VLSI Sign Process Syst Sign Image Video Technol

View full text Add to dashboard Cite

Designing embedded systems is a challenging task during which wrong choices can lead to extremely costly re-design loops, especially when these wrong choices are made during the algorithm specification and the mapping over the selected architecture. In this paper we propose a high-level approach for design space exploration, using a usual standard language as input. More precisely we present the two first steps of the Design Trotter framework: (i) the specification step and its underlying internal model (HCDFG: Hierarchical and Control Data Flow Graph) and (ii) the characterization step which takes place very early in the design flow. Indeed, once transformed into our internal representation, the specification is rapidly and automatically characterized and explored at the algorithmic level. The framework provides the designer with metrics so that he can evaluate, very early in the design process, the impact of algorithmic choices on resource requirements in terms of processing, control, memory bandwidth and potential parallelism at different levels of granularity. The overall aim of our approach is to improve the algorithm/architecture matching that sorely influences the implementation efficiency in terms of silicon area, performances and energy consumption. We give examples which illustrate how designers can refer to the outcomes of the Design Trotter framework in order to select or build suitable architectures for specific applications.

show abstract

Section: Overview Of the Design Trotter Frameworkmentioning

confidence: 99%

Section: Overview Of the Design Trotter Frameworkmentioning

confidence: 99%

Section: Characterizationmentioning

confidence: 99%

Section: Memory Modelmentioning

confidence: 99%

See 2 more Smart Citations

Algorithmic-level Specification and Characterization of Embedded Multimedia Applications with Design Trotter

Moullec

Diguet

Amor

et al. 2006

J VLSI Sign Process Syst Sign Image Video Technol

View full text Add to dashboard Cite

show abstract

“…The minimum unfolding factor n  is computed with the maximum distance of inter-iteration dependency dmax [12]. The loop core is duplicated (n  times) and the ASAP dates of the duplicated graphs are shifted according to the value of dmax.…”

Section: Adaptive Dfg and Cdfg-loop Schedulingmentioning

confidence: 99%