The ability for providing a hardware platform which can be customized on a per-application basis under software control has established the emphReconfigurable Computing (mathcalRC) as a new computing paradigm. A machine employing the mathcalRC paradigm is referred to as a emphField-Programmable Custom Computing Machine (FCCM). So far the FCCMs have been classified according to implementation criteria. In this presentation we propose a new classification of the FCCMs according to architectural criteria. Specifically, to analyze the phenomena inside FCCMs, we introduce a new formalism based on microcode, in which any reconfigurable operation is executed as a microprogrammed sequence with two basic stages: ∼attSET, and ∼attEXECUTE. We also introduce the concepts of spatial and temporal constructions to describe the structure of set of the reconfigurable units. The ∼attSET/EXECUTE formalism and the spatial/temporal constructions are the two most important criteria we use in building the taxonomy of field-programmable custom computing machines. © 2008 Springer.
CITATION STYLE
Sima, M., Vassiliadis, S., & Cotofana, S. (2008). A taxonomy of field-programmable custom computing machines an architectural approach. In Fine-and Coarse-Grain Reconfigurable Computing (pp. 299–378). Springer Netherlands. https://doi.org/10.1007/978-1-4020-6505-7_7
Mendeley helps you to discover research relevant for your work.