Achieving Efficient Routing in Reconfigurable DCNs

Heavy and highly dynamic traffic demands in today's data center networks (DCNs) pose great challenges to efficient traffic engineering. With gigabit bandwidth, wireless communication technologies, such as free space optics and 60GHz wireless, are promising to augment DCNs and enable efficient traffic engineering. Complementary to the emerging reconfigurable architectures, we aim to achieve efficient routing and effectively balance the load with the performance guarantee. We derive a general interference model and propose a decomposition technique with proven performance guarantee and solve the load balancing problem in reconfigurable DCNs. In addition, we propose two solutions, WiRo and OFS, to flexibly reconfigure network topology and enable hybrid-routing with paths consisting of both stable wired links and flexible wireless links with different methods. Our measurement-facilitated and trace-driven simulations demonstrate that our solutions outperform existing flow scheduling algorithms with the average throughput of large flows increased by up to 190% and the average completion time reduced by up to 72.6%. Meanwhile, the average completion time of small flows is reduced by up to 64.5%.

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

[2]  Minming Li,et al.  Towards Maximal Service Profit in Geo-Distributed Clouds , 2019, 2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS).

[3]  Stefan Schmid,et al.  Efficient Non-Segregated Routing for Reconfigurable Demand-Aware Networks , 2019, 2019 IFIP Networking Conference (IFIP Networking).

[4]  Konstantina Papagiannaki,et al.  c-Through: part-time optics in data centers , 2010, SIGCOMM '10.

[5]  T. S. Eugene Ng,et al.  A Tale of Two Topologies: Exploring Convertible Data Center Network Architectures with Flat-tree , 2017, SIGCOMM.

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

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

[8]  Alex C. Snoeren,et al.  Inside the Social Network's (Datacenter) Network , 2015, Comput. Commun. Rev..

[9]  Stefan Schmid,et al.  Toward demand-aware networking: a theory for self-adjusting networks , 2018, CCRV.

[10]  Peng-Jun Wan,et al.  Multiflows in multihop wireless networks , 2009, MobiHoc '09.

[11]  Stefan Schmid,et al.  Demand-Aware Network Designs of Bounded Degree , 2017, DISC.

[12]  Lili Qiu,et al.  Impact of Interference on Multi-Hop Wireless Network Performance , 2003, MobiCom '03.

[13]  Albert G. Greenberg,et al.  Data center TCP (DCTCP) , 2010, SIGCOMM '10.

[14]  Haitao Wu,et al.  Per-packet load-balanced, low-latency routing for clos-based data center networks , 2013, CoNEXT.

[15]  Xiuzhen Cheng,et al.  Dynamic Scheduling for Wireless Data Center Networks , 2013, IEEE Transactions on Parallel and Distributed Systems.

[16]  J. Y. Yen,et al.  Finding the K Shortest Loopless Paths in a Network , 2007 .

[17]  Desh Ranjan,et al.  Balls and bins: A study in negative dependence , 1996, Random Struct. Algorithms.

[18]  Stefan Schmid,et al.  Characterizing the algorithmic complexity of reconfigurable data center architectures , 2018, ANCS.

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

[20]  Michael J. Freedman,et al.  Scalable, optimal flow routing in datacenters via local link balancing , 2013, CoNEXT.

[21]  David S. Johnson,et al.  Computers and Intractability: A Guide to the Theory of NP-Completeness , 1978 .

[22]  Kai Han,et al.  RUSH: Routing and scheduling for hybrid data center networks , 2015, 2015 IEEE Conference on Computer Communications (INFOCOM).

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

[24]  Srikanth Kandula,et al.  Dynamic load balancing without packet reordering , 2007, CCRV.

[25]  Christian E. Hopps,et al.  Analysis of an Equal-Cost Multi-Path Algorithm , 2000, RFC.

[26]  Albert G. Greenberg,et al.  VL2: a scalable and flexible data center network , 2009, SIGCOMM '09.

[27]  Christian Scheideler,et al.  SplayNet: Towards Locally Self-Adjusting Networks , 2016, IEEE/ACM Transactions on Networking.

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

[29]  Kai Chen,et al.  Neural Network Meets DCN , 2018, Proc. ACM Meas. Anal. Comput. Syst..

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

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

[32]  Robert D. Carr,et al.  Randomized metarounding , 2002, Random Struct. Algorithms.

[33]  Yin Tat Lee,et al.  An Almost-Linear-Time Algorithm for Approximate Max Flow in Undirected Graphs, and its Multicommodity Generalizations , 2013, SODA.

[34]  Ao Tang,et al.  HALO: Hop-by-Hop Adaptive Link-State Optimal Routing , 2015, IEEE/ACM Transactions on Networking.

[35]  Stefan Schmid,et al.  Demand-Aware Network Design with Minimal Congestion and Route Lengths , 2019, IEEE INFOCOM 2019 - IEEE Conference on Computer Communications.

[36]  Keqiang He,et al.  Presto: Edge-based Load Balancing for Fast Datacenter Networks , 2015, SIGCOMM.

[37]  Prabhakar Raghavan,et al.  Randomized rounding: A technique for provably good algorithms and algorithmic proofs , 1985, Comb..

[38]  Min Zhu,et al.  WCMP: weighted cost multipathing for improved fairness in data centers , 2014, EuroSys '14.

[39]  Ben Y. Zhao,et al.  Cutting the cord: a robust wireless facilities network for data centers , 2014, MobiCom.

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

[41]  Ming Zhang,et al.  MicroTE: fine grained traffic engineering for data centers , 2011, CoNEXT '11.

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

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

[44]  Rong Pan,et al.  Let It Flow: Resilient Asymmetric Load Balancing with Flowlet Switching , 2017, NSDI.

[45]  Nikhil R. Devanur,et al.  ProjecToR: Agile Reconfigurable Data Center Interconnect , 2016, SIGCOMM.

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

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

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

[49]  Xin Wang,et al.  Traffic-aware virtual machine migration in topology-adaptive DCN , 2016, 2016 IEEE 24th International Conference on Network Protocols (ICNP).