DocumentCode :
3378294
Title :
Unified analytical global placement for large-scale mixed-size circuit designs
Author :
Hsu, Meng-Kai ; Chang, Yao-Wen
Author_Institution :
Grad. Inst. of Electron. Eng., Nat. Taiwan Univ., Taipei, Taiwan
fYear :
2010
fDate :
7-11 Nov. 2010
Firstpage :
657
Lastpage :
662
Abstract :
A modern chip often contains large numbers of pre-designed macros (e.g., embedded memories, IP blocks) and standard cells, with very different sizes. The fast-growing design complexity with large-scale mixed-size macros and standard cells has caused significant challenges to modern circuit placement. Analytical algorithms have been shown to be most effective for standard-cell placement, but the problems with the rotation and legalization of large macros impose intrinsic limitations for analytical placement. Consequently, most recent works on mixed-size placement resort to combinatorial macro placement. Instead, this paper presents the first attempt to resolve the intrinsic problems with a unified analytical approach. Unlike traditional analytical placement that uses only wire and density forces to optimize the positions of circuit components, we present a new force, the rotation force, to handle macro orientation for analytical mixed-size placement. The rotation force tries to rotate each macro to its desired orientation based on the wire connections on this macro. A cross potential model is also proposed to increase the rotation freedom during placement. The final orientation of each macro with legalization consideration is then determined by mathematical programming at the end of global placement. Experimental results show the effectiveness and efficiency of our approach. Compared with state-of-the-art mixed-size placement approaches (such as FLOP, CG, and MP-tree), our approach achieves the best average wirelength efficiently.
Keywords :
circuit complexity; large-scale systems; mathematical programming; network synthesis; analytical global placement; circuit placement; combinatorial macro placement; cross potential model; design complexity; large-scale mixed-size circuit designs; large-scale mixed-size macros; mathematical programming; rotation force; Force; Law; Mathematical model; Mathematical programming; Pins; Wire;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer-Aided Design (ICCAD), 2010 IEEE/ACM International Conference on
Conference_Location :
San Jose, CA
ISSN :
1092-3152
Print_ISBN :
978-1-4244-8193-4
Type :
conf
DOI :
10.1109/ICCAD.2010.5654240
Filename :
5654240
Link To Document :
بازگشت