The performance prediction model: a methodology for estimating the performance of an FPGA implementation of an algorithm
Tobias Jeger, Rolf Enzler, Didier Cottet, Gerhard Tröster · Repository for Publications and Research Data (ETH Zurich) · 2000
Employing FPGAs for computing purposes requires a mapping process.Although this procedure saves substantial development time compared to the physical design of an ASIC, several mappings are almost always needed until the constraints can be met.The Performance Prediction Model (PPM) is a means to help reducing the development time while providing an estimation of the performance to be achieved.The estimation is based on a preliminary implementation 1 only and separates different influences, which improves its portability.The PPM yields an accuracy of about 10% in average and is thus suitable for finding a good implementation for starting the actual mapping process.