Casey Reardon, Brian Holland, Alan D. George, Greg Stitt, Herman Lam
2012 ACM Transactions on Embedded Computing Systems  
Reconfigurable computing (RC) is emerging as a promising area for embedded computing, where complex systems must balance performance, flexibility, cost, and power. The difficulty associated with RC development suggests improved strategic planning and analysis techniques can save significant development time and effort. This article presents a new abstract modeling language and environment, the RC Modeling Language (RCML), to facilitate efficient design-space exploration of RC systems at the
more » ... mation modeling level, i.e. before building a functional implementation. Two integrated analysis tools and case studies, one analytical and one simulative, are presented illustrating relatively accurate automated analysis of systems modeled in RCML. tail to meaningfully represent candidate designs. Therefore, an abstract estimation model is needed at the beginning of the development process allowing systems to be modeled and analyzed quickly and meaningfully before moving towards a more detailed implementation. Designers can repeatedly analyze and alter the abstract model based on feedback from analysis tools, which can save considerable time and cost in overall development by reducing wasted effort in later stages of development. This article introduces and presents a new language and environment tailored for estimation modeling of RC systems and applications, called the Reconfigurable Computing Modeling Language (RCML). RCML is designed to allow users to efficiently model RC systems in the early stages of RC development. The RCML framework enables users to separately model the algorithm and the execution platform architecture under study, providing specific constructs for defining parallelism, communication patterns, and other common aspects in RC applications. A complete system model is then created by mapping an application and platform model together. This process allows individual models to be reused for any number of mappings, supporting efficient iterative design-space exploration early in the development process.
doi:10.1145/2331147.2331153 fatcat:5vobgex6ubh6hoh4pojzfepfiq