TY - JOUR
T1 - Packaging and Interconnect Considerations in Neuromorphic Photonic Accelerators
AU - Nezami, Mohammadreza Sanadgol
AU - De Lima, Thomas Ferreira
AU - Mitchell, Matthew
AU - Yu, Shangxuan
AU - Wang, Jing
AU - Bilodeau, Simon
AU - Zhang, Weipeng
AU - Al-Qadasi, Mohammed
AU - Taghavi, Iman
AU - Tofini, Alexander
AU - Lin, Stephen
AU - Shastri, Bhavin J.
AU - Prucnal, Paul R.
AU - Chrostowski, Lukas
AU - Shekhar, Sudip
N1 - Publisher Copyright:
© 1995-2012 IEEE.
PY - 2023
Y1 - 2023
N2 - Developing compute platforms capable of performing computations at high speed is essential for data processing in the next generation of data centers and edge devices. A neuromorphic photonic accelerator on a silicon photonic platform is a promising solution. Compared to silicon photonic data communication transceiver modules, neuromorphic photonic accelerators constitute a large number of active and passive components and optoelectronic devices to handle the parallel processing. Thus, an increased number of optical and electrical interconnects are required, making the packaging of such processors challenging. Moreover, thermal and electrical crosstalk can dramatically degrade the performance of such processors. Thus, packaging a neuromorphic photonic accelerator for efficient processing and data movement requires careful considerations at the chip, module, and board levels. This work investigates the challenges and potential solutions for optical coupling, optical and electrical interconnections, processor-memory communication, and thermal and electrical cross-talk to develop neuromorphic photonic accelerators.
AB - Developing compute platforms capable of performing computations at high speed is essential for data processing in the next generation of data centers and edge devices. A neuromorphic photonic accelerator on a silicon photonic platform is a promising solution. Compared to silicon photonic data communication transceiver modules, neuromorphic photonic accelerators constitute a large number of active and passive components and optoelectronic devices to handle the parallel processing. Thus, an increased number of optical and electrical interconnects are required, making the packaging of such processors challenging. Moreover, thermal and electrical crosstalk can dramatically degrade the performance of such processors. Thus, packaging a neuromorphic photonic accelerator for efficient processing and data movement requires careful considerations at the chip, module, and board levels. This work investigates the challenges and potential solutions for optical coupling, optical and electrical interconnections, processor-memory communication, and thermal and electrical cross-talk to develop neuromorphic photonic accelerators.
KW - Heterogeneous integration
KW - co-packaging
KW - optical computing
KW - silicon photonics
UR - http://www.scopus.com/inward/record.url?scp=85137548993&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85137548993&partnerID=8YFLogxK
U2 - 10.1109/JSTQE.2022.3200604
DO - 10.1109/JSTQE.2022.3200604
M3 - Article
AN - SCOPUS:85137548993
SN - 1077-260X
VL - 29
JO - IEEE Journal of Selected Topics in Quantum Electronics
JF - IEEE Journal of Selected Topics in Quantum Electronics
IS - 2
M1 - 6100311
ER -