Extending the Power-Efficiency and Performance of Photonic Interconnects for Heterogeneous Multicores with Machine Learning

• Published in: 2018 IEEE International Symposium on High Performance Computer Architecture (HPCA) pp. 480 - 491
• Main Authors: Van Winkle, Scott, Kodi, Avinash Karanth, Bunescu, Razvan, Louri, Ahmed
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.02.2018
• Subjects: Bandwidth; Graphics processing units; Machine Learning; Multicore processing; Network on Chips; Photonics; Power lasers; Power Scaling; System-on-chip
• ISSN: 2378-203X
• DOI: 10.1109/HPCA.2018.00048

As communication energy exceeds computation energy in future technologies, traditional on-chip electrical interconnects face fundamental challenges in the many-core era. Photonic interconnects have been proposed as a disruptive technology solution due to superior performance per Watt, distance independent energy consumption and CMOS compatibility for on-chip interconnects. Static power due to the laser being always switched on, varying link utilization due to spatial and temporal traffic fluctuations and thermal sensitivity are some of the critical challenges facing photonics interconnects. In this paper, we propose photonic interconnects for heterogeneous multicores using a checkerboard pattern that clusters CPU-GPU cores together and implements bandwidth reconfiguration using local router information without global coordination. To reduce the static power, we also propose a dynamic laser scaling technique that predicts the power level for the next epoch using the buffer occupancy of previous epoch. To further improve power-performance trade-offs, we also propose a regression-based machine learning technique for scaling the power of the photonic link. Our simulation results demonstrate a 34% performance improvement over a baseline electrical CMESH while consuming 25% less energy per bit when dynamically reallocating bandwidth. When dynamically scaling laser power, our buffer-based reactive and ML-based proactive prediction techniques show 40 - 65% in power savings with 0 - 14% in throughput loss depending on the reservation window size.