A framework for reconfigurable computing: task scheduling and context management

Abstract

Dynamically reconfigurable architectures are emerging as a viable design alternative to implement a wide range of computationally intensive applications. At the same time, an urgent necessity has arisen for support tool development to automate the design process and achieve optimal exploitation of the architectural features of the system. Task scheduling and context (configuration) management become very critical issues in achieving the high performance that digital signal processing (DSP) and multimedia applications demand. This article proposes a strategy to automate the design process which considers all possible optimizations that can be carried out at compilation time, regarding context and data transfers. This strategy is general in nature and could be applied to different reconfigurable systems. We also discuss the key aspects of the scheduling problem in a reconfigurable architecture such as MorphoSys. In particular, we focus on a task scheduling methodology for DSP and multimedia applications, as well as the context management and scheduling optimizations.

Power-performance trade-offs for reconfigurable computing

“…The reconfiguration latency has to be minimized in order to improve application performance. Temporal partitioning [8] and multi-context scheduling [5] techniques can be used to minimize this penalty.…”

A design flow tailored for self dynamic reconfigurable architecture

“…The model adopted is similar to the one proposed in [35]. The idea is to iterate the execution of each module a certain number of times, and in order to obtain modules whose running time is comparable to the reconfiguration time of other modules, thus hiding reconfiguration overhead, as shown in Figure 2.…”

Maximum edge matching for reconfigurable computing

“…The following steps describe an iteration t of the ACO algorithm to solve the maximum edge matching (3,6) problem:…”

“…Due to a lack of area the corresponding modules on a device have to be reconfigured. Many authors consider task scheduling and prefetching techniques to hide reconfiguration time from the application [4,5,2,8,6]. However, reduced configuration data and power can be achieved by the use of custom reconfigurable architectures [3,14] or by advanced implementation methods that exploit similar properties of tasks.…”