%0 Journal Article
%T Evaluation of Runtime Task Mapping Using the rSesame Framework
%A Kamana Sigdel
%A Carlo Galuzzi
%A Koen Bertels
%A Mark Thompson
%A Andy D. Pimentel
%J International Journal of Reconfigurable Computing
%D 2012
%I Hindawi Publishing Corporation
%R 10.1155/2012/234230
%X 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. 1. Introduction In recent years, reconfigurable architectures [1, 2] have received an increasing attention due to their adaptability and short time to market. Reconfigurable architectures use reconfigurable hardware, such as Field Programmable Gate Array (FPGA) [3, 4] or other programmable hardware (e.g., Complex Programmable Logic Device (CPLD) [5], reconfigurable Datapath Array (rDPA) [6]). These hardware resources are frequently coupled with a core processor, typically a General Purpose Processor (GPP), which is responsible for controlling the reconfigurable hardware. Part of the application¡¯s tasks is executed on the GPP, while the rest of the tasks are executed on the hardware. In general, the hardware implementation of an application is more efficient in terms of performance than a software implementation. As a result, reconfigurable architectures enhance the whole application through an implementation of selected application kernels onto the reconfigurable hardware, while preserving the flexibility of the software execution with the GPP at the same time [7, 8]. The design of such architectures is subject to numerous design constraints and requirements, such as performance, chip area, power consumption, and memory. As a consequence, the design of heterogeneous reconfigurable systems imposes several challenges to system designers such as hardware-software partitioning, Design Space Exploration (DSE), task mapping, and task scheduling. Reconfigurable systems can evolve
%U http://www.hindawi.com/journals/ijrc/2012/234230/