An insertion loss balance aware routing scheme in photonic network on chip

Authors:
Zhijuan Chang;Jianxiong Tang;Yaohui Jin
Affiliations:
State Key Lab of Advanced Optical Communication System and Network , Shanghai Jiao Tong University, Shanghai, China;State Key Lab of Advanced Optical Communication System and Network , Shanghai Jiao Tong University, Shanghai, China;State Key Lab of Advanced Optical Communication System and Network , Shanghai Jiao Tong University, Shanghai, China
Venue:
ICICS'09 Proceedings of the 7th international conference on Information, communications and signal processing
Year:
2009

Citing 6
Cited 0

The turn model for adaptive routing

Journal of the ACM (JACM)
The Odd-Even Turn Model for Adaptive Routing

IEEE Transactions on Parallel and Distributed Systems
On the Design of a Photonic Network-on-Chip

NOCS '07 Proceedings of the First International Symposium on Networks-on-Chip
NoC-Based FPGA: Architecture and Routing

NOCS '07 Proceedings of the First International Symposium on Networks-on-Chip
Photonic NoC for DMA Communications in Chip Multiprocessors

HOTI '07 Proceedings of the 15th Annual IEEE Symposium on High-Performance Interconnects
Link-load balance aware mapping and routing for NoC

WSEAS Transactions on Circuits and Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

The paper presents a novel energy efficient routing scheme for photonic network on chip (PNoC) based on the micro-resonator which takes the insertion loss and the path-setup latency into consideration in order to design a high performance per watt multi-processors. The path-setup latency is a major contributor to the whole system's latency that has something to do with queuing. And insertion loss which has impact on power consumption is closely related to the number of micro-resonators in on state. To solve the multi-objective optimization problem, we design a straight line first heuristic routing scheme. The goal of this routing scheme is to reduce the path-setup latency and insertion loss. Simulation results show that this scheme can low the latency with lower the insertion loss as little as possible.