Measuring and utilising the correlation between signal connectivity and signal positioning for FPGAs containing multi-bit building blocks

As the logic capacity of field-programmable gate arrays (FPGAs) increases, there has been a corresponding increase in the variety of FPGA building blocks. From a mere collection of conventional logic blocks, FPGAs can now include digital signal processors, multipliers, multi-bit addressable memory cells and even processor cores. One of the common characteristics of these new building blocks is their multi-bit design, where each block is designed specifically to process several bits of data at a time. This multi-bit processing paradigm is significantly different from the single-bit processing design of the conventional FPGA logic blocks, as it creates differentiation in signals through its bussed structures. Consequently, the correlation between the positions of the signals in buses and the connectivity of these signals is examined. On the basis of correlation measurements, a multi-bit routing architecture is then proposed along with its routing tool. It is experimentally shown that, compared with the conventional routing architectures, the multi-bit architecture requires 6-12% less area to implement. In particular, it needs 27% fewer routing switches to connect its multi-bit blocks to their routing tracks and 18% less configuration memory to store the configuration information.

[1]  Vaughn Betz,et al.  A fast routability-driven router for FPGAs , 1998, FPGA '98.

[2]  Jonathan Rose,et al.  Synthesizing datapath circuits for FPGAs with emphasis on area minimization , 2002, 2002 IEEE International Conference on Field-Programmable Technology, 2002. (FPT). Proceedings..

[3]  Jonathan Rose,et al.  Measuring and utilizing the correlation between signal connectivity and signal positioning for FPGAs containing multi-bit building blocks , 2005, International Conference on Field Programmable Logic and Applications, 2005..

[4]  Vaughn Betz,et al.  How Much Logic Should Go in an FPGA Logic Block? , 1998, IEEE Des. Test Comput..

[5]  Vaughn Betz,et al.  Architecture and CAD for Deep-Submicron FPGAS , 1999, The Springer International Series in Engineering and Computer Science.

[6]  Jan M. Rabaey,et al.  A reconfigurable data-driven multiprocessor architecture for rapid prototyping of high throughput DSP algorithms , 1993, [1993] Proceedings of the Twenty-sixth Hawaii International Conference on System Sciences.

[7]  Jonathan Rose,et al.  The effect of LUT and cluster size on deep-submicron FPGA performance and density , 2004 .

[8]  Vivek Sarkar,et al.  Baring It All to Software: Raw Machines , 1997, Computer.

[9]  Jürgen Becker,et al.  Architecture and application of a dynamically reconfigurable hardware array for future mobile communication systems , 2000, Proceedings 2000 IEEE Symposium on Field-Programmable Custom Computing Machines (Cat. No.PR00871).

[10]  Carl Ebeling,et al.  Placement and routing tools for the Triptych FPGA , 1995, IEEE Trans. Very Large Scale Integr. Syst..

[11]  Jan M. Rabaey,et al.  A reconfigurable multiprocessor IC for rapid prototyping of algorithmic-specific high-speed DSP data paths , 1992 .

[12]  Jonathan Rose,et al.  Using multi-bit logic blocks and automated packing to improve field-programmable gate array density for implementing datapath circuits , 2004, Proceedings. 2004 IEEE International Conference on Field- Programmable Technology (IEEE Cat. No.04EX921).

[13]  Seth Copen Goldstein,et al.  PipeRench: A Reconfigurable Architecture and Compiler , 2000, Computer.

[14]  Jonathan Rose,et al.  Field-programmable gate array architectures and algorithms optimized for implementing datapath circuits , 2004 .

[15]  Jonathan Rose,et al.  Using bus-based connections to improve field-programmable gate array density for implementing datapath circuits , 2005, FPGA '05.