Advertisement

2-Hop Eclipse: A Fast Algorithm for Bandwidth-Efficient Data Center Switching

  • Liang LiuEmail author
  • Long Gong
  • Sen Yang
  • Jun (Jim) Xu
  • Lance Fortnow
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10967)

Abstract

A hybrid-switched data center network interconnects its racks of servers with a combination of a fast circuit switch that a schedule can reconfigure at significant cost and a much slower packet switch that a schedule can reconfigure at negligible cost. Given a traffic demand matrix between the racks, how can we best compute a good circuit switch configuration schedule that meets most of the traffic demand, leaving as little as possible for the packet switch to handle?

In this paper we propose 2-hop Eclipse, a new hybrid switch scheduling algorithm that strikes a much better tradeoff between the performance of the hybrid switch and the computational complexity of the algorithm, both in theory and in simulations, than the current state of the art solution Eclipse/Eclipse++.

Keywords

Traffic Demand Matrix Hybrid Switching Circuit Switch Packet Switching Switch Scheduling 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Notes

Acknowledgments

This work is supported in part by US NSF through award CNS 1423182.

References

  1. 1.
    DeCusatis, C.: Optical interconnect networks for data communications. J. Lightw. Technol. 32(4), 544–552 (2014)CrossRefGoogle Scholar
  2. 2.
    Patel, P., et al.: Ananta: cloud scale load balancing. In: SIGCOMM Computer Communication Review, vol. 43, pp. 207–218. ACM (2013)CrossRefGoogle Scholar
  3. 3.
    Farrington, N., et al.: Helios: a hybrid electrical/optical switch architecture for modular data centers. SIGCOMM Comput. Commun. Rev. 40(4), 339–350 (2010)CrossRefGoogle Scholar
  4. 4.
    Wang, H., et al.: Design and demonstration of an all-optical hybrid packet and circuit switched network platform for next generation data centers. In: OFC. Optical Society of America (2010)Google Scholar
  5. 5.
    Farrington, N., et al.: A 10 us hybrid optical-circuit/electrical-packet network for datacenters. In: OFC. Optical Society of America (2013)Google Scholar
  6. 6.
    Chen, K., et al.: OSA: an optical switching architecture for data center networks with unprecedented flexibility. IEEE/ACM Trans. Netw. 22(2), 498–511 (2014)CrossRefGoogle Scholar
  7. 7.
    Wang, G., et al.: c-through: part-time optics in data centers. In: SIGCOMM Computer Communication Review, vol. 40, pp. 327–338. ACM (2010)CrossRefGoogle Scholar
  8. 8.
    Liu, H., et al.: Circuit switching under the radar with REACToR. NSDI 14, 1–15 (2014)Google Scholar
  9. 9.
    Porter, G., et al.: Integrating microsecond circuit switching into the data center. SIGCOMM Comput. Commun. Rev. 43(4), 447–458 (2013)CrossRefGoogle Scholar
  10. 10.
    Li, X., et al.: On scheduling optical packet switches with reconfiguration delay. IEEE J. Sel. Areas Commun. 21(7), 1156–1164 (2003)CrossRefGoogle Scholar
  11. 11.
    Bojja Venkatakrishnan, S., et al.: Costly circuits, submodular schedules and approximate carathéodory theorems. In: SIGMETRICS, pp. 75–88. ACM (2016)CrossRefGoogle Scholar
  12. 12.
    Li, C., et al.: Using indirect routing to recover from network traffic scheduling estimation error. In: ACM/IEEE Symposium on Architectures for Networking and Communications Systems (2017)Google Scholar
  13. 13.
    Even, S., et al.: On the complexity of time table and multi-commodity flow problems. In: FOCS, pp. 184–193. IEEE (1975)Google Scholar
  14. 14.
    Ford Jr., L.R., et al.: A suggested computation for maximal multi-commodity network flows. Manage. Sci. 5(1), 97–101 (1958)MathSciNetCrossRefGoogle Scholar
  15. 15.
    Chiang, T.C.: Multi-commodity network flows. Oper. Res. 11(3), 344–360 (1963)CrossRefGoogle Scholar
  16. 16.
    Garg, N., et al.: Faster and simpler algorithms for multicommodity flow and other fractional packing problems. SIAM J. Comput. 37(2), 630–652 (2007)MathSciNetCrossRefGoogle Scholar
  17. 17.
    Bojja Venkatakrishnan, S.: In-Person Discussions at Sigmetrics Conference, June 2017Google Scholar
  18. 18.
    Mekkittikul, A., et al.: A starvation-free algorithm for achieving 100% throughput in an input-queued switch. In: Proceedings of ICCCN, vol. 96, pp. 226–231. Citeseer (1996)Google Scholar
  19. 19.
    Liu, H., et al.: Scheduling techniques for hybrid circuit/packet networks. In: ACM CoNEXT. CoNEXT 2015, pp. 41:1–41:13. ACM, New York. ACM (2015)Google Scholar
  20. 20.
    Duan, R., et al.: A scaling algorithm for maximum weight matching in bipartite graphs. In: SODA, pp. 1413–1424. SIAM (2012)CrossRefGoogle Scholar
  21. 21.
    Benson, T., et al.: Network traffic characteristics of data centers in the wild. In: IMC, pp. 267–280. ACM (2010)Google Scholar
  22. 22.
    Ghobadi, M., et al.: ProjecTOR: agile reconfigurable data center interconnect. In: SIGCOMM, pp. 216–229 (2016)Google Scholar
  23. 23.
    Hamedazimi, N., et al.: FireFLY: a reconfigurable wireless data center fabric using free-space optics. In: SIGCOMM, pp. 319–330 (2014)CrossRefGoogle Scholar
  24. 24.
    Gale, D., et al.: College admissions and the stability of marriage. Am. Math. Mon. 69(1), 9–15 (1962)MathSciNetCrossRefGoogle Scholar
  25. 25.
    Inukai, T.: An efficient SS/TDMA time slot assignment algorithm. IEEE Trans. Commun. 27(10), 1449–1455 (1979)MathSciNetCrossRefGoogle Scholar
  26. 26.
    Towles, B., et al.: Guaranteed scheduling for switches with configuration overhead. IEEE/ACM Trans. Netw. 11(5), 835–847 (2003)CrossRefGoogle Scholar
  27. 27.
    Wu, B., et al.: Nxg05-6: minimum delay scheduling in scalable hybrid electronic/optical packet switches. In: GLOBECOM, pp. 1–5. IEEE (2006)Google Scholar
  28. 28.
    Gopal, I.S., et al.: Minimizing the number of switchings in an SS/TDMA system. IEEE Trans. Commun. 33(6), 497–501 (1985)CrossRefGoogle Scholar
  29. 29.
    Fu, S., et al.: Cost and delay tradeoff in three-stage switch architecture for data center networks. In: HPSR, pp. 56–61. IEEE (2013)Google Scholar
  30. 30.
    Wang, C.-H., et al.: End-to-end scheduling for all-optical data centers. In: INFOCOM, pp. 406–414. IEEE (2015)Google Scholar
  31. 31.
    Wang, C.-H.: et al.: Heavy traffic queue length behavior in switches with reconfiguration delay. arXiv preprint arXiv:1701.05598 (2017)

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Liang Liu
    • 1
    Email author
  • Long Gong
    • 1
  • Sen Yang
    • 1
  • Jun (Jim) Xu
    • 1
  • Lance Fortnow
    • 1
  1. 1.Georgia Institute of TechnologyAtlantaGeorgia

Personalised recommendations