Analysis of power model of multi-core microprocessor architecture

被引:0
|
作者
Chen Z. [1 ]
Liu C. [2 ]
Hou S. [3 ]
Guo Y. [4 ]
机构
[1] Institute of Combat Support, Army Academe, Wuxi
[2] College of Computer Science and Electronic Engineering, Hunan University, Changsha
[3] Department of Basic Courses, Information Engineering University, Luoyang
[4] School of Computer, National University of Defense Technology, Changsha
基金
中国国家自然科学基金;
关键词
Architecture level; Multi-core processor; Peak power consumption; Process simulator;
D O I
10.11817/j.issn.1672-7207.2019.07.014
中图分类号
学科分类号
摘要
The FT-SHSim simulation tool platform was used to model the mainstream microprocessor core models SMT and MSS. Using advanced CMOS technology, the simulation experiment of power consumption evaluation at the architecture level obtained the process requirements of different microprocessor structures and the performance and scale required by the same microprocessor structure during different processes. This provided a reference value for process requirements and implementation methods for the early stages of design, improving quality, shortening cycles, and accelerating design convergence. The results show that in the 22 nm process of the minimum line width, the 128-core SMT processor model has a peak power of 116 W and 64-core MSS processor model has a peak power of 161 W. © 2019, Central South University Press. All right reserved.
引用
收藏
页码:1611 / 1618
页数:7
相关论文
共 20 条
  • [1] Afacan E., Berke Yelten M., Dundar G., Review: analog design methodologies for reliability in nanoscale CMOS circuits, 2017 14th International Conference on Synthesis, Modeling, Analysis and Simulation Methods and Applications to Circuit Design (SMACD), pp. 1-4, (2017)
  • [2] Wang Y., Jin X., Zhou A., Novel LDNMOS embedded SCR with strong ESD robustness based on 0.5 μm 18 V CDMOS technology, Journal of Central South University, 22, 2, pp. 552-559, (2015)
  • [3] Salah K., More than Moore and beyond CMOS: New interconnects schemes and new circuits architectures, 2017 IEEE 19th Electronics Packaging Technology Conference (EPTC), pp. 1-6, (2017)
  • [4] Bhonge S., Boppana V., Low power chips: a fabless asic perspective, Proceeding of the 13th International Symposium on Low Power Electronics and Design (ISLPED'08), pp. 347-348, (2008)
  • [5] Melikyan V., Martirosyan M., Babayan D., Et al., Multi-voltage and multi-threshold low power design techniques for ORCA processor based on 14 nm technology, 2018 IEEE 38th International Conference on Electronics and Nanotechnology (ELNANO), pp. 116-120, (2018)
  • [6] Wang L., Wang D., Way prediction set-associative data cache for low power digital signal processors, 2016 IEEE 13th International Conference on Signal Processing (ICSP), pp. 508-512, (2016)
  • [7] Lakshminarayana N.B., Kim H., Block-precise processors: low-power processors with reduced operand store accesses and result broadcasts, IEEE Transactions on Computers, 64, 11, pp. 3102-3114, (2015)
  • [8] Reimann T., Sze C.C.N., Reis R., Gate sizing and threshold voltage assignment for high performance microprocessor designs, The 20th Asia and South Pacific Design Automation Conference, pp. 214-219, (2015)
  • [9] Campbell S.A., Fabrication Engineering at the Micro- and Nanoscale, pp. 599-612, (2012)
  • [10] Huang P., Xing Z., Impacts of NBTI/PBTI on power gated SRAM, Journal of Central South University, 20, 5, pp. 1298-1306, (2013)