Title :
IR-Drop Aware Clustering Technique for Robust Power Grid in FPGAs
Author :
Kumar, Akhilesh ; Anis, Mohab
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Waterloo, Waterloo, ON, Canada
fDate :
7/1/2011 12:00:00 AM
Abstract :
IR-drop management in the power supply network of a chip is one of the critical design challenges in nanometer VLSI circuits. Techniques developed for application-specific integrated circuits cannot be directly applied for IR drop management in field-programmable gate arrays (FPGAs) because of the programmable nature of FPGAs. This paper proposes a novel clustering technique for improving the supply voltage profile in power grid of FPGAs. The proposed clustering technique not only improves the minimum voltage at any node in the circuit, but also reduces the variance in supply voltage across the nodes in the power grid. Results indicate that a reduction of up to 36% in IR-drop and 27% in spatial Vdd variation can be achieved using the proposed clustering technique.
Keywords :
VLSI; field programmable gate arrays; logic CAD; pattern clustering; power aware computing; power grids; FPGA; IR-drop aware clustering technique; robust power grid; supply voltage; supply voltage profile; Circuits; Design automation; Field programmable gate arrays; Logic gates; Power grids; Power supplies; Robustness; Very large scale integration; Voltage; Wires; Computer-aided design (CAD); field-programmable gate array (FPGA); power grid; reliability; voltage drop;
Journal_Title :
Very Large Scale Integration (VLSI) Systems, IEEE Transactions on
DOI :
10.1109/TVLSI.2010.2047123