Energy-efficient optical Network-on-Chip architecture for heterogeneous multicores

We propose an optical crossbar-based dual-ring Network-on-Chip architecture that connects the last level caches with both CPU and GPU cores to reduce network latency and improve energy-efficiency. Our simulation results show a 17% throughput improvement and a 36% energy reduction over a comparable electronic network.

[1]  Jung Ho Ahn,et al.  Corona: System Implications of Emerging Nanophotonic Technology , 2008, 2008 International Symposium on Computer Architecture.

[2]  David R. Kaeli,et al.  Leveraging Silicon-Photonic NoC for Designing Scalable GPUs , 2015, ICS.

[3]  Ahmed Louri,et al.  Dynamic Reconfiguration of 3D Photonic Networks-on-Chip for Maximizing Performance and Improving Fault Tolerance , 2012, 2012 45th Annual IEEE/ACM International Symposium on Microarchitecture.

[4]  Chen Sun,et al.  DSENT - A Tool Connecting Emerging Photonics with Electronics for Opto-Electronic Networks-on-Chip Modeling , 2012, 2012 IEEE/ACM Sixth International Symposium on Networks-on-Chip.

[5]  David R. Kaeli,et al.  Multi2Sim: A simulation framework for CPU-GPU computing , 2012, 2012 21st International Conference on Parallel Architectures and Compilation Techniques (PACT).