The Journal of Supercomputing

, Volume 71, Issue 4, pp 1249–1276 | Cite as

A cost-efficient congestion management methodology for fat-trees using traffic pattern detection

Article

Abstract

Interconnection networks have a great impact on the performance of parallel systems. These networks provide the communication mechanism and framework needed by parallel applications. One such important network is fat-tree. Selection functions were shown to have a great impact on the performance of fat-trees. Selection functions perform differently under certain traffic patterns. The stage and destination priority (SADP) selection function was shown to have better performance in case of uniform traffic while the stage and origin priority (SAOP) selection function was shown to perform better in case of hot-spot traffic. In this paper, we propose a cost-efficient congestion management mechanism for fat-trees that choose a certain selection function for certain traffic pattern. The mechanism has the ability to detect the current traffic pattern and switch to a certain selection function that is proved to give better performance under the detected traffic pattern. This directly decreases the congestion in the network. First, we analyze the hot-spot traffic in fat-trees if SADP selection function is used. We derive a condition for the existence of hot-spot traffic if SADP function is used. We give an implementation for detecting this condition. Once this condition is detected, the network is forced to switch to use the SAOP selection function. Then, we use the analysis of SAOP to derive a condition to detect that a non hot-spot traffic exists in the fat-tree. We give an implementation for detecting this condition. In turn, we switch back to the SADP selection function. We use synthetic workloads to show the accuracy of the proposed mechanism for detecting the hot-spot traffic in the network. We show that the proposed mechanism incurs a constant number of bits per physical link as an overhead. Finally, we compare the proposed mechanism with other techniques.

References

  1. 1.
    Gilabert F, Gomez ME, Lopez P, Duato J (2006) On the Influence of the selection function on the performance of fat-trees. In: European Conference on Parallel ComputingGoogle Scholar
  2. 2.
    Gomez C, Gilabert F, Gomez ME, Lopez P, Duato J (2007) Deterministic versus adaptive routing in fat-trees. In: Proceedings of workshop on communication architecture on clustersGoogle Scholar
  3. 3.
    Youngsik Kim, Kwon Oh-Young, Han Tack-Don, Youngsong Mun (1997) Design and performance analysis of the practical fat-tree network using a butterfly network. J Syst Archit 43(1–5):355–363Google Scholar
  4. 4.
    DeHon A (1990) A fat-tree routing for transit. A Technical Report, Massachusetts Institute of Technology, Cambridge, MA, USAGoogle Scholar
  5. 5.
  6. 6.
    Zahavi E, Johnson G, Kerbyson DJ, Lang M (2010) Optimized InfiniBandTM fattree routing for shift all-to-all communication patterns. Concurrency and computation: practice and experience (ISSN: 1532–0634) 22(2):217–231. doi:10.1002/cpe.1527
  7. 7.
    Farouk A, El-Boghdadi HM (2011) On the Influence of selection function on the performance of fat-trees under hot-spot traffic. In: Proceedings of the 9th ACS/IEEE International Conference on Computer Systems and Applications (AICCSA) Sharm El-Sheikh, EgyptGoogle Scholar
  8. 8.
    Farouk A, El-Boghdadi HM (2012) A methodology for easing the congestion in fat-trees using traffic pattern detection. In: 10th IEEE International Symposium on Parallel and Distributed Processing with Applications (ISPA 2012). MadridGoogle Scholar
  9. 9.
    Liu JC, Shin KG, Chang C (1992) Prevention of hot-spots in packet-switched multistage interconnection networks. In: International Conference on Par. and Distributed Systems. Hsin Chu, pp 1–6Google Scholar
  10. 10.
    Yang Z, Ma H (2009) Hotspot avoidance for P2P streaming distribution application: a game theoretic approach. IEEE Trans Parallel Distrib Syst 20(2):219–232CrossRefMathSciNetGoogle Scholar
  11. 11.
    Loucif S, Ould-Khaoua M, Geyong M (2007) A queuing model for predicting message latency in uni-directional k-ary N-cubes with deterministic routing and non-uniform traffic. Cluster Comput J 10(2):229–239CrossRefGoogle Scholar
  12. 12.
    Lang Tomás, Kurisaki Lance (1990) Nonuniform traffic spots (NUTS) in multistage interconnection networks. J. Parallel Distrib. Comput. 10(1):55–67CrossRefGoogle Scholar
  13. 13.
    Guay WL, Bogdanski B, Reinemo SA, Lysne O, Skeie T (2011) vFtree—a fat-tree routing algorithm using virtual lanes to alleviate congestion. In: Parallel and Distributed Processing Symposium (IPDPS), 2011 IEEE International. pp 197–208Google Scholar
  14. 14.
    Santoro N, Khatib R (1985) Routing without routing tables. Tech. report SCS-TR-6, School of Computer Science, Carleton University, 1982. Also as: labeling and implicit routing in networks. Comput J 28(1):5–8CrossRefMATHMathSciNetGoogle Scholar
  15. 15.
    Duato J, Johnson I, Flich J, Naven F, García P, Nachiondo T (2005) A new scalable and cost-effective congestion management strategy for lossless multistage interconnection networks. In: Proceedings of the 11th Int’l Symposium on High-Performance Computer Architecture (HPCA-11 2005)Google Scholar
  16. 16.
    Ferrer JL, Baydal E, Robles A, Lopez P, Duato J (2010) A scalable and early congestion management mechanism for MINs. Parallel Distributed and Network-Based Processing. pp 43–50Google Scholar
  17. 17.
    Escudero-Sahuquillo Jesus, Garcia Pedro J, Quiles Francisco J, Flich Jose, Duato Jose (2011) OBQA: smart and cost-efficient queue scheme for head-of-line blocking elimination in fat-trees. J Parallel Distrib Comput 71(11):1460–1472CrossRefGoogle Scholar
  18. 18.
    Escudero-Sahuquillo Jesus, Garcia Pedro J, Quiles Francisco J, Reinemo Sven-Arne, Skeie Tor, Lysne Olav, Duato Jose (2014) A new proposal to deal with congestion in InfiniBand-based fat-trees. J Parallel Distrib Comput 74(1):1802–1819CrossRefGoogle Scholar
  19. 19.
    Bayar S, Yurdakul A (2012) A dynamically reconfigurable communication architecture for multicore embedded systems. J Syst Archit 58(3—-4):140–159CrossRefGoogle Scholar
  20. 20.
    Arabnia HR, Smith JW (1993) A reconfigurable interconnection network for imaging operations and its implementation using a multi-stage switching box. In: Proceedings of the 7th annual international high performance computing conference. The 1993 high performance computing: new horizons supercomputing symposium. Canada, pp 349–357Google Scholar
  21. 21.
    Bhandarkar SM, Arabnia HR (1995) The REFINE multiprocessor—theoretical properties and algorithms. Parallel Comput 21(11):1783–1805CrossRefGoogle Scholar
  22. 22.
    Gran EG, Reinemo SA, Lysne O, Skeie T, Zahavi E, Shainer G (2012) Exploring the scope of the InfiniBand congestion control mechanism. In: Parallel Distributed Processing Symposium (IPDPS), 2012 IEEE 26th International. pp 1131–1143Google Scholar
  23. 23.
    Duato J, Yalamanchili S, Ni L (2004) Interconnection networks. An engineering approach. Morgan KaufmannGoogle Scholar
  24. 24.

Copyright information

© Springer Science+Business Media New York 2015

Authors and Affiliations

  1. 1.Computer Engineering Department, Faculty of EngineeringCairo UniversityGizaEgypt

Personalised recommendations