A framework for partitioning computational intensive applications in hybrid reconfigurable platforms

In this paper, we describe a prototype software framework that implements a formalized methodology for partitioning computational intensive applications between reconfigurable hardware blocks of different granularity. A hybrid granularity reconfigurable generic architecture is considered for this methodology, so as the methodology is applicable to a large variety of hybrid reconfigurable architectures. Although, the proposed framework is parametrical in respect to the mapping procedures to the fine- and coarse-grain reconfigurable units, we provide mapping algorithms for these types of hardware. The experimental results show the effectiveness of the functionality partitioning framework. We have validated the framework using two real-world applications, an OFDM transmitter and a JPEG encoder. For the OFDM transmitter, a maximum clock cycles decrease of 82% relative to an all fine-grain mapping solution is achieved. The performance improvement for the JPEG is 44%.