PowerSyn: A Logic Synthesis Framework With Early Power Optimization

被引:3
作者
Zou, Sunan [1 ,2 ]
Zhang, Jiaxi [1 ,2 ]
Shi, Bizhao [1 ,2 ]
Luo, Guojie [1 ,2 ]
机构
[1] Peking Univ, Sch Comp Sci, Natl Key Lab Multimedia Informat Proc, Beijing 100871, Peoples R China
[2] Peking Univ, Ctr Energy Efficient Comp & Applicat, Beijing 100871, Peoples R China
基金
中国国家自然科学基金;
关键词
Optimization; Power demand; Switches; Integrated circuit modeling; Estimation; Logic gates; Feature extraction; Logic synthesis; power modeling and estimation; synthesis for low power; technology mapping; FPGA; MINIMIZATION;
D O I
10.1109/TCAD.2023.3297069
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Power is a great concern in integrated circuits (ICs) design flow, especially in portable devices. As an early stage in electronic design automation (EDA), logic synthesis can significantly affect the quality of the design. It is essential to optimize power in logic synthesis. However, logic synthesis only has a limited concern in power due to its inaccurate estimation. This is because critical physical information is missing at this stage. Furthermore, the empirical optimization sequences need enhancement, and they are not optimal for power, while optimizing power in the early stage is effective. Technology mapping can also improve power optimization with comprehensive power metrics in this sub-15 nm era. Therefore, we propose PowerSyn, a logic synthesis framework with early power optimization. It consists of a practical power model, a power-oriented logic optimization module, and a technology mapping stage. The power model leverages probability propagation considering glitches and static power. The acrlong RL-based logic optimization generates high-quality and rapid-convergence command sequence with early power optimization. We also modify traditional technology mapping with novel power-related metrics. We evaluate PowerSyn on the EPFL benchmark suite. Experiment results show that our flow achieves an average power savings of 16.1% compared to the state-of-the-art open-source logic optimization flow. It also delivers an 8.8% and a 2.1% reduction in latency and area, respectively. The flow incurs less than 12.2% execution time overhead during inference for command generation.
引用
收藏
页码:203 / 216
页数:14
相关论文
共 52 条
[41]   FlowTune: End-to-End Automatic Logic Optimization Exploration via Domain-Specific Multiarmed Bandit [J].
Neto, Walter Lau ;
Li, Yingjie ;
Gaillardon, Pierre-Emmanuel ;
Yu, Cunxi .
IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, 2023, 42 (06) :1912-1925
[42]  
Schulman J, 2017, Arxiv, DOI arXiv:1707.06347
[43]  
Shum W., 2011, 2011 International Symposium on Low Power Electronics and Design (ISLPED 2011), P27, DOI 10.1109/ISLPED.2011.5993599
[44]  
Sutton RS, 2000, ADV NEUR IN, V12, P1057
[45]  
Wolf C., Yosys Open SYnthesis Suite
[46]  
YANG S, 1991, LOGIC SYNTHESIS OPTI
[47]   Developing Synthesis Flows Without Human Knowledge [J].
Yu, Cunxi ;
Xiao, Houping ;
De Micheli, Giovanni .
2018 55TH ACM/ESDA/IEEE DESIGN AUTOMATION CONFERENCE (DAC), 2018,
[48]  
Zhang J., 2021, PROC INT C COMPUT AI, P1
[49]   GRANNITE: Graph Neural Network Inference for Transferable Power Estimation [J].
Zhang, Yanqing ;
Ren, Haoxing ;
Khailany, Brucek .
PROCEEDINGS OF THE 2020 57TH ACM/EDAC/IEEE DESIGN AUTOMATION CONFERENCE (DAC), 2020,
[50]  
Zhang Zhiru., 2015, IPSJ TRANSACTION SYS, V8, P12