Thor: A Scalable Hybrid Switching Architecture for Data Centers

Optical interconnects are emerging as a high bandwidth and low energy alternative to traditional electrical networks. However, most exiting designs put optical switching in the core layer due to issues, including limited scalability, potential bottleneck of the control plane, and high cost of the multi-wavelength switch. To this end, we propose Thor, a hybrid network architecture using optical interconnects as the main load bearing portion. It employs the hypercube topology and multi-hop circuit switching to enable server-level optical access, thus solving the scalability and connectivity limitations. Then, in order to build a faster control plane operating short-lived circuits in a large scale, Thor uses a distributed control system over the electrical network, with a new path setup mechanism capable of establishing multiple optical circuits concurrently. Moreover, Thor employs a novel optical switch design to utilize multiple wavelengths more efficiently. Our evaluation results show that Thor consumes at least 59% and 6% less power than existing electrical and optical designs. For performance, Thor is able to deliver 90% bisection bandwidth of a non-blocking network and reduce the end-to-end latency significantly. Finally, by separately delivering mice and elephant flows, Thor achieves salient improvements in average flow completion times compared with existing approaches.

[1]  Konstantina Papagiannaki,et al.  c-Through: part-time optics in data centers , 2010, SIGCOMM 2010.

[2]  S. Johnsson,et al.  Spanning balanced trees in Boolean cubes , 1989 .

[3]  Theodore R. Bashkow,et al.  A large scale, homogeneous, fully distributed parallel machine, I , 1977, ISCA '77.

[4]  Ankit Singla,et al.  OSA: An Optical Switching Architecture for Data Center Networks With Unprecedented Flexibility , 2012, IEEE/ACM Transactions on Networking.

[5]  Kostas Katrinis,et al.  Performance evaluation of hybrid optical switch architecture for data center networks , 2016, Opt. Switch. Netw..

[6]  Amin Vahdat,et al.  Jupiter Rising: A Decade of Clos Topologies and Centralized Control in Google's Datacenter Network , 2015, Comput. Commun. Rev..

[7]  Amin Vahdat,et al.  Hedera: Dynamic Flow Scheduling for Data Center Networks , 2010, NSDI.

[8]  Amin Vahdat,et al.  Helios: a hybrid electrical/optical switch architecture for modular data centers , 2010, SIGCOMM '10.

[9]  S. Lennart Johnsson,et al.  Optimum Broadcasting and Personalized Communication in Hypercubes , 1989, IEEE Trans. Computers.

[10]  Sujata Banerjee,et al.  ElasticTree: Saving Energy in Data Center Networks , 2010, NSDI.

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

[12]  Bo Qin,et al.  Towards cost-effective and low latency data center network architecture , 2016, Comput. Commun..

[13]  Ion Stoica,et al.  A cost comparison of datacenter network architectures , 2010, CoNEXT.

[14]  Wei Bai,et al.  Information-Agnostic Flow Scheduling for Commodity Data Centers , 2015, NSDI.

[15]  Qunfeng Dong,et al.  WaveCube: A scalable, fault-tolerant, high-performance optical data center architecture , 2015, 2015 IEEE Conference on Computer Communications (INFOCOM).

[16]  Randy H. Katz,et al.  DeTail: reducing the flow completion time tail in datacenter networks , 2012, SIGCOMM '12.

[17]  William J. Dally,et al.  Flattened butterfly: a cost-efficient topology for high-radix networks , 2007, ISCA '07.

[18]  Hong Liu,et al.  Energy proportional datacenter networks , 2010, ISCA.

[19]  Andrew W. Moore,et al.  Characterizing 10 Gbps network interface energy consumption , 2010, IEEE Local Computer Network Conference.

[20]  Kostas Katrinis,et al.  HOSA: hybrid optical switch architecture for data center networks , 2015, Conf. Computing Frontiers.

[21]  Baochun Li,et al.  RepFlow: Minimizing flow completion times with replicated flows in data centers , 2013, IEEE INFOCOM 2014 - IEEE Conference on Computer Communications.

[22]  Yu Gong,et al.  Passive optical interconnects at top of the rack: offering high energy efficiency for datacenters. , 2015, Optics express.

[23]  Lei Shi,et al.  Dcell: a scalable and fault-tolerant network structure for data centers , 2008, SIGCOMM '08.

[24]  David A. Maltz,et al.  Network traffic characteristics of data centers in the wild , 2010, IMC '10.

[25]  Kostas Katrinis,et al.  Performance evaluation of a hybrid optical/electrical interconnect , 2015, IEEE/OSA Journal of Optical Communications and Networking.

[26]  Tajana Simunic,et al.  Dynamic optical switching for latency sensitive applications , 2015, 2015 ACM/IEEE Symposium on Architectures for Networking and Communications Systems (ANCS).

[27]  Amin Vahdat,et al.  Scale-Out Networking in the Data Center , 2010, IEEE Micro.

[28]  Roberto Proietti,et al.  DOS - A scalable optical switch for datacenters , 2010, 2010 ACM/IEEE Symposium on Architectures for Networking and Communications Systems (ANCS).

[29]  Nick McKeown,et al.  pFabric: minimal near-optimal datacenter transport , 2013, SIGCOMM.

[30]  Roberto Proietti,et al.  A Scalable, Low-Latency, High-Throughput, Optical Interconnect Architecture Based on Arrayed Waveguide Grating Routers , 2015, Journal of Lightwave Technology.

[31]  Amin Vahdat,et al.  Integrating microsecond circuit switching into the data center , 2013, SIGCOMM.

[32]  Alex C. Snoeren,et al.  RotorNet: A Scalable, Low-complexity, Optical Datacenter Network , 2017, SIGCOMM.

[33]  Amin Vahdat,et al.  A scalable, commodity data center network architecture , 2008, SIGCOMM '08.

[34]  Haitao Wu,et al.  BCube: a high performance, server-centric network architecture for modular data centers , 2009, SIGCOMM '09.