Wireless Data Center Networks: Advances, Challenges, and Opportunities

Data center networks (DCNs) are essential infrastructures to embrace the era of highly diversified massive amount of data generated by emerging technological applications. In order to store and process such a data deluge, today's DCNs have to deploy enormous length of wires to interconnect a plethora of servers and switches. Unfortunately, wired DCNs with uniform and inflexible link capacities expose several drawbacks such as high cabling cost and complexity, low space utilization, and lack of bandwidth efficiency. Wireless DCNs (WDCNs) have emerged as a promising solution to reduce the time, effort, and cost spent on deploying and maintaining the wires. Thanks to its reconfigurability and flexibility, WDCNs can deliver higher throughputs by efficiently utilizing the bandwidth and mitigate the chronic DCN problems of oversubscription and hotspots. Moreover, wireless links enhance the fault-tolerance and energy efficiency by eliminating the need for error-prone power-hungry switches. Accordingly, this paper first compares virtues and drawbacks of millimeter wave (mmWave), terahertz (THz), and optical wireless communications as potential candidates. Thereafter, an in-depth discussion on advances and challenges in WDCNs is provided including physical and virtual topology design, quality of service (QoS) provisioning, flow classification, data grooming, and load balancing. Finally, exciting research opportunities are presented to promote the prospects of WDCNs.

[1]  Nasir Saeed,et al.  Underwater Optical Wireless Communications, Networking, and Localization: A Survey , 2018, Ad Hoc Networks.

[2]  Albert G. Greenberg,et al.  The nature of data center traffic: measurements & analysis , 2009, IMC '09.

[3]  Jitender S. Deogun,et al.  OWCell: Optical wireless cellular data center network architecture , 2017, 2017 IEEE International Conference on Communications (ICC).

[4]  Deke Guo,et al.  Design and optimization of VLC based small-world data centers , 2018, Frontiers of Computer Science.

[5]  Ben Y. Zhao,et al.  Mirror mirror on the ceiling: flexible wireless links for data centers , 2012, SIGCOMM.

[6]  George Varghese,et al.  CONGA: distributed congestion-aware load balancing for datacenters , 2015, SIGCOMM.

[7]  Mohamed-Slim Alouini,et al.  Design and provisioning of optical wireless data center networks: A traffic grooming approach , 2018, 2018 IEEE Wireless Communications and Networking Conference (WCNC).

[8]  Himanshu Shah,et al.  FireFly , 2014, SIGCOMM.

[9]  Paramvir Bahl,et al.  Flyways To De-Congest Data Center Networks , 2009, HotNets.

[10]  Mohamed-Slim Alouini,et al.  Optical wireless data center networks: potentials, limitations, and prospects , 2019, OPTO.

[11]  Dezun Dong,et al.  FlyCast: Free-Space Optics Accelerating Multicast Communications in Physical Layer , 2015, Comput. Commun. Rev..

[12]  Ming Zhang,et al.  Understanding data center traffic characteristics , 2010, CCRV.

[13]  Darko Kirovski,et al.  On the Feasibility of Completely Wirelesss Datacenters , 2013, IEEE/ACM Transactions on Networking.

[14]  Xiang-Yang Li,et al.  Diamond: Nesting the Data Center Network With Wireless Rings in 3-D Space , 2016, IEEE/ACM Transactions on Networking.

[15]  Mohamed-Slim Alouini,et al.  LightFD: A Lightweight Flow Detection Mechanism for Traffic Grooming in Optical Wireless DCNs , 2018, 2018 IEEE Global Communications Conference (GLOBECOM).

[16]  Paramvir Bahl,et al.  Augmenting data center networks with multi-gigabit wireless links , 2011, SIGCOMM 2011.

[17]  Nei Kato,et al.  A Failure-Tolerant and Spectrum-Efficient Wireless Data Center Network Design for Improving Performance of Big Data Mining , 2015, 2015 IEEE 81st Vehicular Technology Conference (VTC Spring).