Abstract

The exponential growth in emerging applications in datacenters leads to significant increases in traffic and bandwidth consumption in datacenter networks. Fine-grained optical switching technology can provide low-latency end-to-end connection with flexible granularity in intra-datacenter networks. In such a scheme, the centralized network controller will compute an end-to-end path and provide switch state commands for each connection request. However, as network scale and traffic requirements continually increase, the centralized controller will face a significant challenge in low-latency end-to-end connection provisioning under a heavy traffic load. Limited by the controller’s computation and processing performance, the continuously arriving connection requests may block the server’s message queue and slow down the controller’s processing rate. In this paper, we propose a multi-controller control scheme for fine-grained all-optical datacenter networks. We take optical time slice switching as an example of fine-grained all-optical switching to illustrate our detailed control scheme. Partitioning rules for the network and controllers’ responsibilities, the routing algorithm, and the signaling process are also discussed. The proposed multi-controller control scheme can greatly reduce each controller’s computation and processing burden and improve network performance. Dynamic datacenter network emulations are launched to verify our proposed scheme. The emulation results indicate that the multi-controller scheme can reduce connection setup delays and request and data failure rates, while the single controller fails to handle all the requests.

© 2018 Optical Society of America

Inter-Domain Optimization and Orchestration for Optical Datacenter Networks

G. Landi, M. Capitani, A. Kretsis, K. Kontodimas, P. Kokkinos, D. Gallico, M. Biancani, K. Christodoulopoulos, and E. Varvarigos
J. Opt. Commun. Netw. 10(7) B140-B151 (2018)

Flexible low-latency metro-access converged network architecture based on optical time slice switching

Jialong Li, Nan Hua, Zhizhen Zhong, Yufang Yu, Xiaoping Zheng, and Bingkun Zhou
J. Opt. Commun. Netw. 11(12) 624-635 (2019)

Scalable and low server-to-server latency data center network architecture based on optical packet inter-rack and intra-rack switching

Georgios Drainakis, Peristera Baziana, and Adonis Bogris
J. Opt. Commun. Netw. 15(11) 804-819 (2023)

Multi-layer resource scheduling architecture and algorithm for a service-oriented optical network based on a fine grain OTN

Yang Zhao, Yi Lin, Yunbo Li, Dechao Zhang, Yucong Liu, Yu Zheng, Dong Wang, Sheng Liu, Shan Cao, Haoyu Feng, Han Li, and Xiang Liu
J. Opt. Commun. Netw. 16(10) F13-F25 (2024)

HiFOST: A Scalable and Low-Latency Hybrid Data Center Network Architecture Based on Flow-Controlled Fast Optical Switches

Fulong Yan, Xuwei Xue, and Nicola Calabretta
J. Opt. Commun. Netw. 10(7) B1-B14 (2018)