Matrix computations (3rd ed.)
Proceedings of the 37th Annual Design Automation Conference
Extended Krylov subspace method for reduced order analysis of linear circuits with multiple sources
Proceedings of the 37th Annual Design Automation Conference
Fast power/ground network optimization based on equivalent circuit modeling
Proceedings of the 38th annual Design Automation Conference
Efficient large-scale power grid analysis based on preconditioned krylov-subspace iterative methods
Proceedings of the 38th annual Design Automation Conference
Proceedings of the 39th annual Design Automation Conference
Proceedings of the 2001 IEEE/ACM international conference on Computer-aided design
Power network analysis using an adaptive algebraic multigrid approach
Proceedings of the 40th annual Design Automation Conference
Power grid reduction based on algebraic multigrid principles
Proceedings of the 40th annual Design Automation Conference
Hierarchical analysis of power distribution networks
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Empire: an efficient and compact multiple-parameterized model order reduction method
Proceedings of the 2007 international symposium on Physical design
A novel technique for incremental analysis of on-chip power distribution networks
Proceedings of the 2007 IEEE/ACM international conference on Computer-aided design
Electromigration study of power-gated grids
Proceedings of the 14th ACM/IEEE international symposium on Low power electronics and design
Electromigration for microarchitects
ACM Computing Surveys (CSUR)
EMPIRE: an efficient and compact multiple-parameterized model-order reduction
IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Hi-index | 0.00 |
This paper proposes a novel hierarchical approach to the efficient analysis of large VLSI power/ground grids. Different from the existing hierarchical approach where sub-circuit equivalent models are sparsified with computation-intensive integer programming and the resulting modeling may lead to larger errors if the top circuit matrix has large condition number, the new approach employs an iterative (relaxation) procedure to explicitly compensate the errors and avoid introducing dense matrix caused by the circuit reduction. We also propose an efficient scheme for partitioning high performance center-bumped P/G grids. Experimental results demonstrate that the new algorithm is more accurate than the existing hierarchical method while delivering more speedup over the flat simulators.