ACDC: An Accuracy- and Congestion-aware Dynamic Traffic Control Method for Networks-on-Chip
Siyuan Xiao1,a, Xiaohang Wang1,b, Maurizio Palesi2, Amit Kumar Singh3 and Terrence Mak4
1South China University of Technology
asyxiao1337@gmail.com
bxiaohangwang@scut.edu.cn
2University of Catania
maurizio.palesi@dieei.unict.it
3University of Essex
a.k.singh@essex.ac.uk
4University of Southampton
t.mak@soton.ac.uk
ABSTRACT
Many applications exhibit error forgiving features. For these applications, approximate computing provides the opportunity of accelerating the execution time or reducing power consumption, by mitigating computation effort to get an approximate result. Among the components on a chip, network-on-chip (NoC) contributes a large portion to system power and performance. In this paper, we exploit the opportunity of aggressively reducing network congestion and latency by selectively dropping data. Essentially, the importance of the dropped data is measured based on a quality model. An optimization problem is formulated to minimize the network congestion with constraint of the result quality. A lightweight online algorithm is proposed to solve this problem. Experiments show that on average, our proposed method can reduce the execution time by as much as 12.87% and energy consumption by 12.42% under strict quality requirement, speed up execution by 19.59% and reduce energy consumption by 21.20% under relaxed requirement, compared to a recent work on approximate computing approach for NoCs.
Keywords: Approximate computing, Many-core system, Onchip network.
