Advertisement

Exceeding the Performance of Two-Tier Fat-Tree: Equality Network Topology

  • Chane-Yuan YangEmail author
  • Chi-Hsiu Liang
  • Hong-Lin Wu
  • Chun-Ho Cheng
  • Chao-Chin Li
  • Chun-Ming Chen
  • Po-Lin Huang
  • Chi-Chuan Hwang
Conference paper
Part of the Lecture Notes in Networks and Systems book series (LNNS, volume 70)

Abstract

A high-performance interconnect topology system named Equality is introduced in this paper for general purpose applications including supercomputing, data center, cloud service, and industrial cluster solutions. Equality is designed based on chordal ring networks. It advances previous discussed chordal ring topologies by a set of systematic linking strategies and routing rules. The Equality topology can be used construct low diameter networks with reasonably low router radices. Equality interconnects are highly symmetric and hence cabling rule and routing logic are simple. Compared with other networks, the Equality topology is flexible in total number of routers, where any even number is allowed. This paper introduces the evaluation of Equality performance using open-source BookSim 2.0 package. The benchmarks of ten traffic models for systems constructed using 36- and 48-port switch are presented to assess the network performance, compared with the very popular 2-tier fat-tree structure. The results show that the Equality networks are resilient in the scenario similar to practical computation.

Keywords

Equality network topology Fat-tree High speed computation Data center BookSim 

References

  1. 1.
    Denning, P.J., Lewis, T.G.: Exponential laws of computing growth. Commun. ACM. 60(1), 54–65 (2016)CrossRefGoogle Scholar
  2. 2.
    Kessler, R.E., Schwarzmeier, J.L.: CRAY T3D: a new dimension for Cray Research. In: Compcon Spring’93, Digest of Papers, pp. 176–182, 22 Feb 1993. IEEEGoogle Scholar
  3. 3.
    Ajima, Y., Sumimoto, S., Shimizu, T.: Tofu: a 6D mesh/torus interconnect for exascale computers. Computer 42(11), 36–41 (2009)CrossRefGoogle Scholar
  4. 4.
    Kim, J., Dally, W.J., Abts, D.: Flattened butterfly: a cost-efficient topology for high-radix networks. In: ACM SIGARCH Computer Architecture News, vol. 35, no. 2, pp. 126–137, 9 June 2007. ACMCrossRefGoogle Scholar
  5. 5.
    Kim, J., Dally, W.J., Scott, S., Abts, D.: Technology-driven, highly-scalable dragonfly topology. In: ACM SIGARCH Computer Architecture News, vol. 36, no. 3, pp. 77–88, 21 June 2008. IEEE Computer SocietyCrossRefGoogle Scholar
  6. 6.
    Valerio, M., Moser, L.E., Melliar-Smith, P.M.: Fault-tolerant orthogonal fat-trees as interconnection networks. In: IEEE First International Conference on Algorithms and Architectures for Parallel Processing, 1995. ICAPP 95. IEEE First ICA/sup 3/PP., vol. 2, pp. 749–754, 19 Apr 1995. IEEEGoogle Scholar
  7. 7.
    Besta, M., Hoefler, T.: Slim fly: a cost effective low-diameter network topology. In: International Conference for High Performance Computing, Networking, Storage and Analysis, SC14, pp. 348–359, 16 Nov 2014. IEEEGoogle Scholar
  8. 8.
    Liang, C.-H., Yang, C.-Y., Li, C.-C., Cheng, C.-H., Wu, Y.-C., Chen, C.-M., Huang, P.-L., Hwang, C.-C.: Equality: a flexible topology system ready for high performance computing. IEEE Trans. Parallel Distrib. Syst. (2018) (to be confirmed)Google Scholar
  9. 9.
    Leiserson, C.E.: Fat-trees: universal networks for hardware-efficient supercomputing. IEEE Trans. Comput. 100(10), 892–901 (1985)CrossRefGoogle Scholar
  10. 10.
    Al-Fares, M., et al.: A scalable, commodity data center network architecture. In: ACM SIGCOMM Computer Communication Review, vol. 38Google Scholar
  11. 11.
    Introducing data center fabric, the next-generation Facebook data center network. Andreyev (2014) [Online]. Available: https://code.facebook.com/posts/360346274145943/
  12. 12.
    Klenk, B., Fröning, H.: An overview of MPI characteristics of exascale proxy applications. In: International Supercomputing Conference. Springer, Cham (2017)Google Scholar
  13. 13.
    Jiang, N., et al.: A detailed and flexible cycle-accurate network-on-chip simulator. In: 2013 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)Google Scholar
  14. 14.
    Pfister, G.F., Norton, V.A.: “Hot spot” contention and combining in multistage interconnection networks. IEEE Trans. Comput. C-34 (10) (1985)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Chane-Yuan Yang
    • 1
    Email author
  • Chi-Hsiu Liang
    • 1
  • Hong-Lin Wu
    • 1
  • Chun-Ho Cheng
    • 1
  • Chao-Chin Li
    • 1
  • Chun-Ming Chen
    • 1
  • Po-Lin Huang
    • 1
  • Chi-Chuan Hwang
    • 1
  1. 1.Department of Engineering ScienceNational Cheng Kung UniversityTainanTaiwan

Personalised recommendations