Abstract

This paper proposes a machine-learning (ML)-aided cognitive approach for effective bandwidth reconfiguration in optically interconnected datacenter/high-performance computing (HPC) systems. The proposed approach relies on a Hyper-X-like architecture augmented with flexible-bandwidth photonic interconnections at large scales using a hierarchical intra/inter-POD photonic switching layout. We first formulate the problem of the connectivity graph and routing scheme optimization as a mixed-integer linear programming model. A two-phase heuristic algorithm and a joint optimization approach are devised to solve the problem with low time complexity. Then, we propose an ML-based end-to-end performance estimator design to assist the network control plane with intelligent decision making for bandwidth reconfiguration. Numerical simulations using traffic distribution profiles extracted from HPC applications traces as well as random traffic matrices verify the accuracy performance of the ML design estimator (${\lt}9\%$ error) and demonstrate up to $5 \times$ throughput gain from the proposed approach compared with the baseline Hyper-X network using fixed all-to-all intra/inter-portable data center interconnects.

© 2021 Optical Society of America

Full Article  |  PDF Article
More Like This
Hierarchical and reconfigurable optical/electrical interconnection network for high-performance computing

Zuoqing Zhao, Bingli Guo, Yu Shang, and Shanguo Huang
J. Opt. Commun. Netw. 12(3) 50-61 (2020)

Virtual Network Reconfiguration for Reducing Energy Consumption in Optical Data Centers

Yuya Tarutani, Yuichi Ohsita, and Masayuki Murata
J. Opt. Commun. Netw. 6(10) 925-942 (2014)

Flexspander: augmenting expander networks in high-performance systems with optical bandwidth steering

Min Yee Teh, Zhenguo Wu, and Keren Bergman
J. Opt. Commun. Netw. 12(4) B44-B54 (2020)

References

You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Figures (6)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Tables (4)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Equations (21)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription