The Community for Technology Leaders
2013 12th IEEE International Conference on Trust, Security and Privacy in Computing and Communications (2013)
Melbourne, VIC Australia
July 16, 2013 to July 18, 2013
pp: 879-886
Shuai Zhang , State Key Lab. of Comput. Archit., Inst. of Comput. Technol., Beijing, China
Zhiyong Liu , State Key Lab. of Comput. Archit., Inst. of Comput. Technol., Beijing, China
Dongrui Fan , State Key Lab. of Comput. Archit., Inst. of Comput. Technol., Beijing, China
Fonglong Song , State Key Lab. of Comput. Archit., Inst. of Comput. Technol., Beijing, China
Mingzhe Zhang , State Key Lab. of Comput. Archit., Inst. of Comput. Technol., Beijing, China
ABSTRACT
This paper discusses energy-performance trade-off of networks-on-chip with real parallel applications. First, we propose an accurate energy-performance analytical model that conduct and analyze the impacts of both frequency-independent and frequency-dependent power. Second, we put together the communication overhead, memory access overhead, frequency scaling, and core count scaling to quantify the performance and energy consumed by NoCs. Third, we propose a new energy-performance optimization method, by choosing a pair of frequency and core count to get optimal energy or performance. Finally, we implement eight PARSEC parallel applications to evaluate our model and the optimization method. The experiment result confirms that our model predicts NoCs energy and performance well, and selects correct frequency level and core count for most parallel applications.
INDEX TERMS
Optimization, Analytical models, Multicore processing, Load modeling, Mathematical model, Telecommunication traffic, Threshold voltage
CITATION

Shuai Zhang, Zhiyong Liu, Dongrui Fan, Fonglong Song and Mingzhe Zhang, "Energy-Performance Modeling and Optimization of Parallel Computing in On-Chip Networks," 2013 12th IEEE International Conference on Trust, Security and Privacy in Computing and Communications(TRUSTCOM), Melbourne, VIC Australia, 2014, pp. 879-886.
doi:10.1109/TrustCom.2013.107
244 ms
(Ver 3.3 (11022016))