Mapping streaming applications on a reconfigurable MPSoC platform at run-time

Hölzenspies, P.K.F.; Smit, Gerard J. M.; Kuper, Jan

doi:10.1109/issoc.2007.4427443

Cited by 9 publications

(6 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A self-edge is added to each new actor to enforce the sequentiality of tokens transmission. A more complex model is proposed by Holzenspies et al [4]: each original channel is replaced by a cycle composed of three actors representing receiving, processing and transmission of tokens. The first and the last actor make reading and writing delays explicit, refining the analyzed model.…”

Section: Related Workmentioning

confidence: 99%

“…For the sake of simplicity the approach proposed in [3] is applied to model the delay introduced by a communication. More complex approaches like the ones proposed in [4], [9] can however be integrated in the proposed methodology. Given a channel c i = {s i , t i , p i , d i , i i }, if s i and t i are mapped on different processing elements, c i is removed from the SDF graph and replaced with two channels and one actor.…”

Section: Communication Modelingmentioning

confidence: 99%

“…To overcome this limitation, several methodologies (e.g., [3], [4]) have been proposed which detail inter-actors communications by adding new actors. These methodologies however focus only on the delay of the single communications and they do not correctly model communication interactions overestimating the throughput.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Modeling Resolution of Resources Contention in Synchronous Data Flow Graphs

Lattuada

Ferrandi

2014

J Sign Process Syst

View full text Add to dashboard Cite

Synchronous Data Flow graphs are widely adopted in the designing of streaming applications, but were originally formulated to describe only how an application is partitioned and which data are exchanged among different tasks. Since Synchronous Data Flow graphs are often used to describe and evaluate complete design solutions, missing information (e.g., mapping, scheduling, etc.) has to be included in them by means of further actors and channels to obtain accurate evaluations. To address this issue preserving the simplicity of the representation, techniques that model data transfer delays by means of ad-hoc actors have been proposed, but they model independently each communication ignoring contentions. Moreover, they do not usually consider at all delays due to buffer contentions, potentially overestimating the throughput of a design solution.In this paper a technique to extend Synchronous Data Flow graphs by adding ad-hoc actors and channels to model resolution of resources contentions is proposed. The results show that the number of added actors and channels is limited but that they can significantly increase the Synchronous Data Flow graph accuracy.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Communication Modelingmentioning

confidence: 99%

See 1 more Smart Citation

Modeling Resolution of Resources Contention in Synchronous Data Flow Graphs

Lattuada

Ferrandi

2014

J Sign Process Syst

View full text Add to dashboard Cite

show abstract

“…The mutex loops added to the flexible filter model allow arbitrary order of execution of the filters, depending on when they are enabled. The technique of modeling buffer space with the addition of back edges and tokens to the SDF has been also applied in several previous works including [Poplavko et al 2003;Hölzenspies et al 2007;Moreira and Bekooij 2007].…”

Section: Related Workmentioning

confidence: 99%

Flexible filters in stream programs

Collins

Carloni

2013

ACM Trans. Embed. Comput. Syst.

View full text Add to dashboard Cite

The stream-processing model is a natural fit for multicore systems because it exposes the inherent locality and concurrency of a program and highlights its separable tasks for efficient parallel implementations. We present flexible filters, a load-balancing optimization technique for stream programs. Flexible filters utilize the programmability of the cores in order to improve the data-processing throughput of individual bottleneck tasks by "borrowing" resources from neighbors in the stream. Our technique is distributed and scalable because all runtime load-balancing decisions are based on point-to-point handshake signals exchanged between neighboring cores. Load balancing with flexible filters increases the system-level processing throughput of stream applications, particularly those with large dynamic variations in the computational load of their tasks. We empirically evaluate flexible filters in a homogeneous multicore environment over a suite of five real-word stream programs.

show abstract

“…The approach presented in [28] consists of a mapper, which determines a mapping of application(s) to an architecture, using a library at runtime. The approach proposed by authors in [29] performs mapping of streaming applications, with real-time requirements, onto a reconfigurable MPSoC architecture. In the same way, Faruque et al [30] present a scheme for runtime-agent-based distributed application mapping for on-chip communication for adaptive NoC-based heterogeneous multiprocessor systems.…”

Section: Related Workmentioning

confidence: 99%

Evaluation of Runtime Task Mapping Using the rSesame Framework

Sigdel

Galuzzi

Bertels

et al. 2012

International Journal of Reconfigurable Computing

View full text Add to dashboard Cite

Performing runtime evaluation together with design time exploration enables a system to be more efficient in terms of various design constraints, such as performance, chip area, and power consumption. rSesame is a generic modeling and simulation framework, which can explore and evaluate reconfigurable systems at both design time and runtime. In this paper, we use the rSesame framework to perform a thorough evaluation (at design time and at runtime) of various task mapping heuristics from the state of the art. An extended Motion-JPEG (MJPEG) application is mapped, using the different heuristics, on a reconfigurable architecture, where different Field Programmable Gate Array (FPGA) resources and various nonfunctional design parameters, such as the execution time, the number of reconfigurations, the area usage, reusability efficiency, and other parameters, are taken into consideration. The experimental results suggest that such an extensive evaluation can provide a useful insight both into the characteristics of the reconfigurable architecture and on the efficiency of the task mapping.

show abstract

Mapping streaming applications on a reconfigurable MPSoC platform at run-time

Cited by 9 publications

References 7 publications

Modeling Resolution of Resources Contention in Synchronous Data Flow Graphs

Modeling Resolution of Resources Contention in Synchronous Data Flow Graphs

Flexible filters in stream programs

Evaluation of Runtime Task Mapping Using the rSesame Framework

Contact Info

Product

Resources

About