System-level power-performance tradeoffs for reconfigurable computing

Abstract

In this paper, we propose a configuration-aware datapartitioning approach for reconfigurable computing. We show how the reconfiguration overhead impacts the data-partitioning process. Moreover, we explore the system-level power-performance tradeoffs available when implementing streaming embedded applications on fine-grained reconfigurable architectures. For a certain group of streaming applications, we show that an efficient hardware/software partitioning algorithm is required when targeting low power. However, if the application objective is performance, then we propose the use of dynamically reconfigurable architectures. We propose a design methodology that adapts the architecture and algorithms to the application requirements. The methodology has been proven to work on a real research platform based on Xilinx devices. Finally, we have applied our methodology and algorithms to the case study of image sharpening, which is required nowadays in digital cameras and mobile phones.Index Terms-Hardware/software (HW/SW) codesign, powerperformance tradeoffs, reconfigurable computing (RC).

Task Scheduling for Context Minimization in Dynamically Reconfigurable Platforms

“…Ghiasi et al [13] proposed an efficient optimal algorithm to minimize the run-time reconfiguration delay in the executions of applications on a dynamically adaptable system under assumptions on several restricted implementation constraints. Noguera and Badia [2,14] introduced a two-version dynamic scheduling algorithm for reconfigurable architectures with or without a prefetching unit. Resano et al [15] proposed a way to revise a given task schedule by considering reconfiguration to minimize the latency overheads.…”