Essential Traffic Parameters for Shared Memory Switch Performance

  • Patrick Eugster
  • Alex Kesselman
  • Kirill Kogan
  • Sergey Nikolenko
  • Alexander Sirotkin
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9439)

Abstract

Cloud applications bring new challenges to the design of network elements, in particular accommodating for the burstiness of traffic workloads. Shared memory switches represent the best candidate architecture to exploit buffer capacity; we analyze the performance of this architecture. Our goal is to explore the impact of additional traffic characteristics such as varying processing requirements and packet values on objective functions. The outcome of this work is a better understanding of the relevant parameters for buffer management to achieve better performance in dynamic environments of data centers. We consider a model that captures more of the properties of the target architecture than previous work and consider several scheduling and buffer management algorithms that are specifically designed to optimize its performance. In particular, we provide analytic guarantees for the throughput performance of our algorithms that are independent from specific distributions of packet arrivals. We furthermore report on a comprehensive simulation study which validates our analytic results.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Aiello, W., Kesselman, A., Mansour, Y.: Competitive buffer management for shared-memory switches. ACM Transactions on Algorithms 5(1) (2008)Google Scholar
  2. 2.
    Aiello, W., Mansour, Y., Rajagopolan, S., Rosén, A.: Competitive queue policies for differentiated services. J. Algorithms 55(2), 113–141 (2005)MathSciNetCrossRefMATHGoogle Scholar
  3. 3.
    Albers, S., Schmidt, M.: On the performance of greedy algorithms in packet buffering. SIAM Journal on Computing 35(2), 278–304 (2005)MathSciNetCrossRefMATHGoogle Scholar
  4. 4.
    Alizadeh, M., Edsall, T., Dharmapurikar, S., Vaidyanathan, R., Chu, K., Fingerhut, A., Lam, V.T., Matus, F., Pan, R., Yadav, N., Varghese, G.: CONGA: distributed congestion-aware load balancing for datacenters. In: ACM SIGCOMM 2014 Conference, pp. 503–514 (2014)Google Scholar
  5. 5.
    Azar, Y., Litichevskey, A.: Maximizing throughput in multi-queue switches. Algorithmica 45(1), 69–90 (2006)MathSciNetCrossRefMATHGoogle Scholar
  6. 6.
    Azar, Y., Richter, Y.: An improved algorithm for CIOQ switches. ACM Transactions on Algorithms 2(2), 282–295 (2006)MathSciNetCrossRefMATHGoogle Scholar
  7. 7.
    BBC News. US Watchdog to Propose New Net Neutrality Rules (2014). http://www.bbc.com/news/technology-27141121.
  8. 8.
    Borodin, A., El-Yaniv, R.: Online Computation and Competitive Analysis. Cambridge University Press (1998)Google Scholar
  9. 9.
    Feng, W.C., Kandlur, D.D., Saha, D., Shin, K.G.: Stochastic fair blue: A queue management algorithm for enforcing fairness. In: INFOCOM, pp. 1520–1529 (2001)Google Scholar
  10. 10.
    Chowdhury, M., Zhong, Y., Stoica, I.: Efficient coflow scheduling with varys. In: SIGCOMM, pp. 443–454 (2014)Google Scholar
  11. 11.
    Chuprikov, P., Nikolenko, S.I., Kogan, K.: Priority queueing with multiple packet characteristics. In: INFOCOM, pp. 1–9 (2015)Google Scholar
  12. 12.
    Costa, P., Donnelly, A., Rowstron, A.I.T., O’Shea, G.: Camdoop: Exploiting in-network aggregation for big data applications. In: Proc. 9th USENIX Symposium on Networked Systems Design and Implementation (NSDI 2012), pp. 29–42 (2012)Google Scholar
  13. 13.
    Das, S., Sankar, R.: Broadcom smart-buffer technology in data center switches for cost-effective performance scaling of cloud applications (2012). https://www.broadcom.com/collateral/etp/SBT-ETP100.pdf
  14. 14.
    Englert, M., Westermann, M.: Lower and upper bounds on FIFO buffer management in QoS switches. Algorithmica 53(4), 523–548 (2009)MathSciNetCrossRefMATHGoogle Scholar
  15. 15.
    Eugster, P., Kogan, K., Nikolenko, S., Sirotkin, A.: Shared memory buffer management for heterogeneous packet processing. In: ICDCS (2014)Google Scholar
  16. 16.
    Floyd, S., Jacobson, V.: Random early detection gateways for congestion avoidance, pp. 397–413 (1993)Google Scholar
  17. 17.
    CAIDA The Cooperative Association for Internet Data Analysis. http://www.caida.org/
  18. 18.
    Goldwasser, M.: A survey of buffer management policies for packet switches. SIGACT News 41(1), 100–128 (2010)CrossRefGoogle Scholar
  19. 19.
    Hong, C.-Y., Kandula, S., Mahajan, R., Zhang, M., Gill, V., Nanduri, M., Wattenhofer, R.: Achieving high utilization with software-driven WAN. In: ACM SIGCOMM 2013 Conference, pp. 15–26 (2013)Google Scholar
  20. 20.
    Jain, S., Kumar, A., Mandal, S., Ong, J., Poutievski, L., Singh, A., Venkata, S., Wanderer, J., Zhou, J., Zhu, M., Zolla, J., Hölzle, U., Stuart, S., Vahdat, A.: B4: experience with a globally-deployed software defined wan. In: ACM SIGCOMM 2013 Conference, pp. 3–14 (2013)Google Scholar
  21. 21.
    Keslassy, I., Kogan, K., Scalosub, G., Segal, M.: Providing performance guarantees in multipass network processors. IEEE/ACM Trans. Netw. 20(6), 1895–1909 (2012)CrossRefGoogle Scholar
  22. 22.
    Kesselman, A., Kogan, K., Segal, M.: Packet mode and QoS algorithms for buffered crossbar switches with FIFO queuing. Distributed Computing 23(3), 163–175 (2010)CrossRefMATHGoogle Scholar
  23. 23.
    Kesselman, A., Kogan, K., Segal, M.: Improved competitive performance bounds for CIOQ switches. Algorithmica 63(1-2), 411–424 (2012)MathSciNetCrossRefMATHGoogle Scholar
  24. 24.
    Kesselman, A., Kogan, K., Segal, M.: Best Effort and Priority Queuing Policies for Buffered Crossbar Switches. Chicago Journal of Theoretical Computer Science (2012)Google Scholar
  25. 25.
    Kesselman, A., Kogan, K.: Nonpreemptive Scheduling of Optical Switches. IEEE Transactions on Communications 55(6), 1212–1219 (2007)CrossRefGoogle Scholar
  26. 26.
    Kesselman, A., Lotker, Z., Mansour, Y., Patt-Shamir, B., Schieber, B., Sviridenko, M.: Buffer overflow management in QoS switches. SIAM Journal on Computing 33(3), 563–583 (2004)MathSciNetCrossRefMATHGoogle Scholar
  27. 27.
    Kesselman, A., Mansour, Y.: Harmonic buffer management policy for shared memory switches. Theor. Comput. Sci. 324(2-3), 161–182 (2004)MathSciNetCrossRefMATHGoogle Scholar
  28. 28.
    Kogan, K., López-Ortiz, A., Nikolenko, S., Scalosub, G., Segal, M.: Large profits or fast gains: A dilemma in maximizing throughput with applications to network processors. CoRR, abs/1202.5755 (2013)Google Scholar
  29. 29.
    Kogan, K., López-Ortiz, A., Nikolenko, S., Sirotkin, A.: Multi-queued network processors for packets with heterogeneous processing requirements. In: COMSNETS, pp. 1–10 (2013)Google Scholar
  30. 30.
    Kogan, K., López-Ortiz, A., Nikolenko, S., Scalosub, G., Segal, M.: Balancing work and size with bounded buffers. In: COMSNETS, pp. 1–8 (2014)Google Scholar
  31. 31.
    Kogan, K., López-Ortiz, A., Nikolenko, S.I., Sirotkin, A.V., Tugaryov, D.: FIFO queueing policies for packets with heterogeneous processing. In: Even, G., Rawitz, D. (eds.) MedAlg 2012. LNCS, vol. 7659, pp. 248–260. Springer, Heidelberg (2012)CrossRefGoogle Scholar
  32. 32.
    Kogan, K., López-Ortiz, A., Nikolenko, S., Sirotkin, A.: A taxonomy of semi-FIFO policies. In: IPCCC, pp. 295–304 (2012)Google Scholar
  33. 33.
    Kogan, K., Nikolenko, S., Keshav, S., López-Ortiz, A.: Efficient demand assignment in multi-connected microgrids with a shared central grid. In: SustainIT, pp. 1–5 (2013)Google Scholar
  34. 34.
    Mansour, Y., Patt-Shamir, B., Lapid, O.: Optimal smoothing schedules for real-time streams. Distributed Computing 17(1), 77–89 (2004)CrossRefMATHGoogle Scholar
  35. 35.
    Nikolenko, S.I., Kogan, K.: Single and multiple buffer processing. In: Encyclopedia of Algorithms. Springer (2015)Google Scholar
  36. 36.
    Sleator, D.D., Tarjan, R.E.: Amortized efficiency of list update and paging rules. Communications of the ACM 28(2), 202–208 (1985)MathSciNetCrossRefGoogle Scholar
  37. 37.
    Yang, H.-C., Dasdan, A., Hsiao, R.-L., Parker Jr., D.S.: Map-reduce-merge: simplified relational data processing on large clusters. In: Proc. ACM SIGMOD International Conference on Management of Data, pp. 1029–1040 (2007)Google Scholar
  38. 38.
    Yu, Y., Gunda, P.K., Isard, M.: Distributed aggregation for data-parallel computing: interfaces and implementations. In: SOSP, pp. 247–260 (2009)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  • Patrick Eugster
    • 1
    • 2
  • Alex Kesselman
    • 3
  • Kirill Kogan
    • 4
  • Sergey Nikolenko
    • 5
    • 6
  • Alexander Sirotkin
    • 7
    • 8
  1. 1.Purdue UniversityLafayetteUSA
  2. 2.Technical University of DarmstadtDarmstadtGermany
  3. 3.Google Inc.Mountain ViewUSA
  4. 4.IMDEA Networks InstituteMadridSpain
  5. 5.National Research University Higher School of EconomicsSt. PetersburgRussia
  6. 6.Steklov Institute of Mathematics at St.PetersburgSt.PetersburgRussia
  7. 7.International Laboratory for Applied Network ResearchNational Research University Higher School of EconomicsMoscowRussia
  8. 8.St. Petersburg Institute for Informatics and Automation of the RASSt. PetersburgRussia

Personalised recommendations