Abstract
This chapter discusses experimental demonstrations of a category of optical switches named low-latency interconnect optical network switches (LIONSs). These switches are based on wavelength routing in arrayed waveguide grating routers (AWGRs). There can be both active and passive implementations for intra- and inter-rack communication in data center and HPC networks. In active LIONS architectures, the switching operation is completely performed in optical domain on the fly without any buffering at the switch. In passive LIONS architectures, the switching operation is performed both in electrical and optical domains. In both passive and active implementations, the unique features of wavelength routing and multiplexing in AWGR allow to build a very low-latency and high-bandwidth switch. First, we demonstrated an active LIONS with loopback buffer scheme (LB-LIONS) and compared the experimental results with networking simulation results. This comparison allowed to validate the correctness of the simulation framework and results. Second, by introducing an all-optical token scheme for optical contention resolution and an all-optical negative acknowledgment technique, we experimentally demonstrated alternative active switch architectures (TOKEN-LIONS and TONAK-LIONS) capable of handling contention resolution in the optical domain while removing the needs of using the loopback buffer as well as the centralized control plane. For what concerns passive LIONS architectures, we carried out experiments for hierarchical all-to-all interconnection between data center servers, racks, and clusters with flexible bandwidth allocation to reduce the congestion created by hot spots in the network.
References
S. Kamei et al., 64 × 64-channel uniform-loss and cyclic-frequency arrayed-waveguide grating router module. Electron. Lett. 39(1), 83–84 (2003)
B. Glance, I.P. Kaminow, R.W. Wilson, Applications of the integrated waveguide grating router. J. Lightwave Technol. 12(6), 957–962 (1994)
Ye, X., et al., DOS – A scalable Optical Switch for Datacenters. ACM/IEEE Symposium on Architectures for Networking and Communications Systems (ANCS), (2010)
R. Proietti et al., Scalable optical interconnect architecture using AWGR-based TONAK LION switch with limited number of wavelengths. J. Lightwave Technol. 31(24), 4087–4097 (2013)
X. Ye et al., Buffering and flow control in optical switches for high performance computing. J. Opt. Commun. Netw. IEEE/OSA 3(8), A59–A72 (2011)
R. Proietti et al., All-optical physical layer NACK in AWGR-based optical interconnects. IEEE Photon. Technol. Lett. 24(5), 410–412 (2012)
R. Proietti et al., Scalable and distributed contention resolution in AWGR-based data center switches using RSOA-based optical mutual exclusion. IEEE J. Sel. Top. Quantum Electron. 19(2), 3600111–3600111 (2013)
R. Proietti et al., An all-optical token technique enabling a fully-distributed control plane in AWGR-based optical interconnects. J. Lightwave Technol. 31(3), 414–422 (2013)
Hubner, M., K. Paulsson, and J. Becker. Parallel and Flexible Multiprocessor System-On-Chip for Adaptive Automotive Applications based on Xilinx MicroBlaze Soft-Cores. in Parallel and Distributed Processing Symposium, 2005. 19th IEEE International. (2005)
Ye, X., et al. DOS – A scalable Optical Switch for Datacenters. in ACM/IEEE Symposium on Architectures for Networking and Communications Systems (ANCS). (2010)
Shin, H.C., et al., Reflective semiconductor optical amplifier. Google Patents, (2006)
Kim, H. and S. Chandrasekhar, Dependence of coherent crosstalk penalty on the OSNR of the signal. Optical Fiber Communication Conference (OFC). (2000)
Proietti, R., et al., Fully-distributed Control Plane by All-Optical Token Technique in AWGR-based Optical Interconnects. J. Lightwave Technol. (2012)
L. Buckman, L. Chen, K. Lau, Crosstalk penalty in all-optical distributed switching networks. IEEE Photon. Technol. Lett. 9(2), 250–252 (1997)
R. Proietti et al., A scalable, low-latency, high-throughput, optical interconnect architecture based on arrayed waveguide grating routers. J. Lightwave Technol. 33(4), 911–920 (2015)
R. Yu et al., A scalable silicon photonic chip-scale optical switch for high performance computing systems. Opt. Express 21(26), 32655–32667 (2013)
P. Grani et al., Photonic interconnects for interposer-based 2.5D/3D integrated systems on a Chip, in Proceedings of the Second International Symposium on Memory Systems, (ACM, Alexandria, 2016), pp. 377–386
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG
About this chapter
Cite this chapter
Proietti, R., Yin, Y., Cao, Z., Nitta, C.J., Akella, V., Ben Yoo, S.J. (2018). Low-Latency Interconnect Optical Network Switch (LIONS). In: Testa, F., Pavesi, L. (eds) Optical Switching in Next Generation Data Centers. Springer, Cham. https://doi.org/10.1007/978-3-319-61052-8_6
Download citation
DOI: https://doi.org/10.1007/978-3-319-61052-8_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-61051-1
Online ISBN: 978-3-319-61052-8
eBook Packages: EngineeringEngineering (R0)