Data Center TCP com Histerese em Switches P4

  • Vítor L. G. Silva UFV
  • José A. M. Nacif UFV
  • Marcos A. M. Vieira UFMG

Resumo


O congestionamento de pacotes em uma rede de computadores causa perda de desempenho no sistema. Esta proposta busca melhorar o desempenho das redes de computadores em centro de dados, especialmente em cenários de congestionamento. O protocolo padrão de controle de congestionamento em centro de dados é o Data Center TCP (DCTCP). Este artigo propõe duas abordagens: i) o uso da técnica de histerese nas filas dos switches para marcar os pacotes, indicando o congestionamento; ii) o retorno de pacotes ao emissor para reduzir a taxa de envio quando houver congestionamento. As abordagens foram implementadas em linguagem P4 utilizando switches P4 programáveis e validado juntamente com o DCTCP. Os resultados mostram que ambas soluções propostas melhoram a vazão da rede.

Referências

Afanasyev, A., Tilley, N., Reiher, P., and Kleinrock, L. (2010). Host-to-host congestion control for tcp. IEEE Communications Surveys & Tutorials, 12(3):304–342.

Alizadeh, M., Greenberg, A., Maltz, D. A., Padhye, J., Patel, P., Prabhakar, B., Sengupta, S., and Sridharan, M. (2010). Data center tcp (dctcp). SIGCOMM Comput. Commun. Rev., 40(4):63–74.

Alizadeh, M., Javanmard, A., and Prabhakar, B. (2011). Analysis of dctcp: Stability, convergence, and fairness. SIGMETRICS Perform. Eval. Rev., 39(1):73–84.

Bas, A. (2020). The bmv2 simple switch target.

Bertotti, G. (1998). Hysteresis in Magnetism: For Physicists, Materials Scientists, and Engineers. Academic Press.

Bosshart, P., Daly, D., Gibb, G., Izzard, M., McKeown, N., Rexford, J., Schlesinger, C., Talayco, D., Vahdat, A., Varghese, G., and Walker, D. (2014). P4: Programming protocol-independent packet processors. SIGCOMM Comput. Commun. Rev., 44(3):87–95.

Cardwell, N., Cheng, Y., Gunn, C. S., Yeganeh, S. H., and Jacobson, V. (2016). Bbr: Congestion-based congestion control. ACM Queue, 14, September-October:20 – 53.

Chen, X., Feibish, S., Koral, Y., Rexford, J., Rottenstreich, O., Monetti, S., and Wang, T.-Y. (2019). Fine-grained queue measurement in the data plane. pages 15–29.

Dong, M., Meng, T., Zarchy, D., Arslan, E., Gilad, Y., Godfrey, B., and Schapira, M. (2018). PCC vivace: Online-learning congestion control. In 15th USENIX Symposium on Networked Systems Design and Implementation (NSDI 18), pages 343–356, Renton, WA. USENIX Association.

Floyd, S. and Jacobson, V. (1993). Random early detection gateways for congestion avoidance. IEEE/ACM Transactions on Networking, 1(4):397–413.

Floyd, S., Ramakrishnan, D. K. K., and Black, D. L. (2001). The Addition of Explicit Congestion Notification (ECN) to IP. RFC 3168.

Geng, J., Yan, J., and Zhang, Y. (2019). P4qcn: Congestion control using p4-capable device in data center networks. Electronics, 8(3):280.

Ha, S., Rhee, I., and Xu, L. (2008). Cubic: a new tcp-friendly high-speed tcp variant. Operating Systems Review, 42:64–74.

Kaur, K., Singh, J., and Ghumman, N. S. (2014). Mininet as software defined networking testing platform. In International Conference on Communication, Computing & Systems (ICCCS), pages 139–42.

Kfoury, E. F., Crichigno, J., and Bou-Harb, E. (2021). An exhaustive survey on p4 programmable data plane switches: Taxonomy, applications, challenges, and future trends.

Malhotra, R., Mandjes, M., Scheinhardt, W., and van den Berg, H. (2009). A feedback uid queue with two congestion control thresholds. Mathematical methods of operations research, 70(1):149–169.

Ramakrishnan, K. K. and Jain, R. (1990). A binary feedback scheme for congestion avoidance in computer networks. ACM Trans. Comput. Syst., 8(2):158–181.

Tirumala, A. (1999). Iperf: The tcp/udp bandwidth measurement tool. http://dast.nlanr.net/Projects/Iperf/.
Publicado
16/08/2021
SILVA, Vítor L. G.; NACIF, José A. M.; VIEIRA, Marcos A. M.. Data Center TCP com Histerese em Switches P4. In: SIMPÓSIO BRASILEIRO DE REDES DE COMPUTADORES E SISTEMAS DISTRIBUÍDOS (SBRC), 39. , 2021, Uberlândia. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2021 . p. 182-195. ISSN 2177-9384. DOI: https://doi.org/10.5753/sbrc.2021.16720.