Performance driven multiway partitioning

Under the interconnect-centric design paradigm, partitioning is seen as the crucial step that defines the interconnect. To meet the performance requirement of today's complex design, performance driven partitioners must consider the amount of interconnect induced by partitioning as well as its impact on performance. In this paper, we provide new performance driven formulation for cell move based top-down multiway partitioning algorithms with consideration of the local and global interconnect delay. In our "constrained acyclic partitioning" formulation, cell moves are restricted to maintain acyclicity in partitioning solution to prevent cyclic dependency among cells in different partitions. In our "relaxed acyclic partitioning" formulation, acyclic constraints are relaxed to give partitioners capability of minimizing cutsize and delay. Our new acyclic constraint based performance driven multiway partitioning algorithm FLARE obtains (i) 4% to 13% better delay compared to the state-of-the-art cutsize minimization based hMetis at almost no increase in cutsize, and (ii) 84% better cutsize compared to the state-of-the-art delay minimization based PRIME at an expense of 16% increase in delay.

[1]  Chung-Kuan Cheng,et al.  A replication cut for two-way partitioning , 1995, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[2]  Chung-Kuan Cheng,et al.  Performance-Driven Partitioning Using a Replication Graph Approach , 1995, 32nd Design Automation Conference.

[3]  Jason Cong,et al.  Large scale circuit partitioning with loose/stable net removal and signal flow based clustering , 1997, ICCAD 1997.

[4]  Majid Sarrafzadeh,et al.  Replication for logic bipartitioning , 1997, 1997 Proceedings of IEEE International Conference on Computer Aided Design (ICCAD).

[5]  Jason Cong,et al.  On Area/Depth Trade-off in LUT-Based FPGA Technology Mapping , 1993, 30th ACM/IEEE Design Automation Conference.

[6]  Eugene L. Lawler,et al.  Module Clustering to Minimize Delay in Digital Networks , 1969, IEEE Transactions on Computers.

[7]  Jason Cong,et al.  Acyclic Multi-Way Partitioning of Boolean Networks , 1994, 31st Design Automation Conference.

[8]  A. Richard Newton,et al.  A cell-replicating approach to minicut-based circuit partitioning , 1991, 1991 IEEE International Conference on Computer-Aided Design Digest of Technical Papers.

[9]  Robert K. Brayton,et al.  On clustering for minimum delay/ara , 1991, 1991 IEEE International Conference on Computer-Aided Design Digest of Technical Papers.

[10]  Jason Cong,et al.  Edge separability based circuit clustering with application to circuit partitioning , 2000, Proceedings 2000. Design Automation Conference. (IEEE Cat. No.00CH37106).

[11]  Honching Li,et al.  Simultaneous circuit partitioning/clustering with retiming for performance optimization , 1999, Proceedings 1999 Design Automation Conference (Cat. No. 99CH36361).

[12]  Jason Cong,et al.  Large scale circuit partitioning with loose/stable net removal and signal flow based clustering , 1997, 1997 Proceedings of IEEE International Conference on Computer Aided Design (ICCAD).

[13]  C. L. Liu,et al.  Optimal clock period clustering for sequential circuits with retiming , 1998, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[14]  Charles M. Fiduccia,et al.  A linear-time heuristic for improving network partitions , 1988, 25 years of DAC.

[15]  Shashi Shekhar,et al.  Multilevel hypergraph partitioning: application in VLSI domain , 1997, DAC.

[16]  Ernest S. Kuh,et al.  Performance-driven system partitioning on multi-chip modules , 1992, [1992] Proceedings 29th ACM/IEEE Design Automation Conference.

[17]  Jason Cong,et al.  An interconnect-centric design flow for nanometer technologies , 1999, 1999 International Symposium on VLSI Technology, Systems, and Applications. Proceedings of Technical Papers. (Cat. No.99TH8453).

[18]  Abbas El Gamal,et al.  Min-cut replication in partitioned networks , 1995, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[19]  Jason Cong,et al.  Performance driven multi-level and multiway partitioning with retiming , 2000, Proceedings 37th Design Automation Conference.

[20]  Martin D. F. Wong,et al.  New algorithms for min-cut replication in partitioned circuits , 1995, ICCAD.

[21]  J. Cong,et al.  Multiway partitioning with pairwise movement , 1998, ICCAD '98.

[22]  Martin D. F. Wong,et al.  Circuit clustering for delay minimization under area and pin constraints , 1995, EDTC '95.