On routing demand and congestion estimation for FPGAs

Interconnection planning is becoming an important design issue for ASICs and large FPGAs. As the technology shrinks and the design density increases, proper planning of routing resources becomes all the more important to ensure rapid and feasible design convergence. One of the most important issues for planning interconnection is the ability to predict the routability of a given placed design. This paper provides insight into the workings of recently proposed method by Lou et al. (2001) and compares it with our proposed methodology, fGREP (2001). We have implemented the two methods for a generic FPGA architecture and compare the performance, accuracy and usability of their estimates. We use the well known FPGA physical design suite VPR (1997), as a common comparison tool. Our experiments show that fGREP produces far better routing estimates but at larger execution times than Lou's method. Insight into what makes the methods work and where they falter are also discussed in detail.