Scalable Models for Autonomous Self-Assembled Reconfigurable Systems

FPGAs are well-suited for applications that need to adjust the composition of computational structures over the lifetime of the application. While the underlying hardware framework for supporting run-time reconfiguration has existed for years, there have been negligibly few FPGA applications that have benefited from this. This is likely not an issue with FPGA architectures, yet it is more likely a formulation problem due to a highly restrictive model for reconfiguration provided by the vendors. This paper introduces a new paradigm in reconfigurable computing, where multiple designs may dynamically compete for resources in the same die. The way to afford this challenge is taking advantage of the benefits of the scalable and adaptable designs, since they achieve a balance between performance and flexibility. Thus, this paper proposes two alternative models for the assembly of computational structures that are conducive to autonomous self-assembled embedded systems, because of their high level of adaptability. To illustrate this new scenario, an example is examined that is based on two competing video image processing tasks.