Reinforcement learning based self-adaptive voltage-swing adjustment of 2.5D I/Os for many-core microprocessor and memory communication
- Resource Type
- Conference
- Authors
- Hantao, Huang; Manoj, P. D. Sai; Xu, Dongjun; Yu, Hao; Hao, Zhigang
- Source
- 2014 IEEE/ACM International Conference on Computer-Aided Design (ICCAD) Computer-Aided Design (ICCAD), 2014 IEEE/ACM International Conference on. :224-229 Nov, 2014
- Subject
- Components, Circuits, Devices and Systems
Engineering Profession
Bit error rate
Receivers
Voltage control
Transmitters
Tuning
Noise
Adaptation models
- Language
- ISSN
- 1092-3152
1558-2434
A reinforcement learning based I/O management is developed for energy-efficient communication between many-core microprocessor and memory. Instead of transmitting data under a fixed large voltage-swing, an online reinforcement Q-learning algorithm is developed to perform a self-adaptive voltage-swing control of 2.5D through-silicon interposer (TSI) I/O circuits. Such a voltage-swing adjustment is formulated as a Markov decision process (MDP) problem solved by model-free reinforcement learning under constraints of both power budget and bit-error-rate (BER). Experimental results show that the adaptive 2.5D TSI I/Os designed in 65nm CMOS can achieve an average of 12.5mw I/O power, 4GHz bandwidth and 3.125pJ/bit energy efficiency for one channel under 10 −6 BER, which has 18.89% power saving and 15.11% improvement of energy efficiency on average.