The effect of LUT and cluster size on deep-submicron FPGA performance and density

Abstract

We use a fully timing-driven experimental flow [4] [15] in which a set of benchmark circuits are synthesized into different cluster-based [2] [3] [15] logic block architectures, which contain groups of LUTs and flip-flops. We look across all architectures with LUT sizes in the range of 2 inputs to 7 inputs, and cluster size from 1 to 10 LUTs. In order to judge the quality of the architecture we do both detailed circuit level design and measure the demand of routing resources for every circuit in each architecture.These experiments have resulted in several key contributions. First, we have experimentally determined the relationship between the number of inputs required for a cluster as a function of the LUT size (K) and cluster size (N). Second, contrary to previous results, we have shown that when the cluster size is greater than four, that smaller LUTs (size 2 and 3) are almost as area efficient as 4-input LUTs, as suggested in [11]. However, our results also show that the performance of FPGAs with these small LUT sizes is significantly worse (by almost a factor of 2) than larger LUTs. Hence, as measured by area-delay product, or by performance, these would be a bad choice. Also, we have discovered that LUT sizes of 5 and 6 produce much better area results than were previously believed. Finally, our results show that a LUT size of 4 to 6 and cluster size of between 4 and 10 provides the best area-delay product for an FPGA.