Should we care about coexistence in L4S? Assessing round-trip time significance and low-latency for all
Resumo
The L4S architecture promises to resolve the throughput-latency trade-off, yet concerns regarding its coexistence with classic TCP persist. The literature presents experimentations on how L4S flows can potentially starve non-L4S ones when deployed on networks with incomplete or no support for L4S, an already known fact. However, few works assess the L4S performance considering networks with full support for it. This work evaluates the dynamics of L4S performance on networks that fully support it and assesses that coexistence is no longer a concern and could also be considered architecturally resolved. We demonstrate that classic flows retain fair throughput with L4S flows but suffer self-inflicted queuing delays exceeding 15 ms, deteriorating their RTT.
Referências
Albisser, O., De Schepper, K., Briscoe, B., Tilmans, O., and Steen, H. (2019). Dualpi2-low latency, low loss and scalable throughput (l4s) aqm. Proc. of Netdev.
Briscoe, B., Schepper, K. D., Bagnulo, M., and White, G. (2023). Low Latency, Low Loss, and Scalable Throughput (L4S) Internet Service: Architecture. RFC 9330.
Cheshire, S. (2025). Ten things you need to know about l4s.
De Schepper, K., Chang, C.-Y., Hamilton, A., and Spruyt, K. (2025). Low Latency Low Loss Scalable Throughput (L4S) for Time-Critical Defense Applications. In 2025 International Conference on Military Communication and Information Systems (ICMCIS), pages 1–6.
Fejes, F., Gombos, G., Laki, S., and Nádasy, S. (2020). On the incompatibility of scalable congestion controls over the internet. In 2020 IFIP Networking Conference (Networking), pages 749–754.
Gettys, J. and Nichols, K. (2012). Bufferbloat: dark buffers in the internet. Commun. ACM, 55(1):57–65.
Graff, P., Marchal, X., Cholez, T., Mathieu, B., Tuffin, S., and Festor, O. (2024). Improving cloud gaming traffic qos: A comparison between class-based queuing policy and l4s. In 2024 8th Network Traffic Measurement and Analysis Conference (TMA), pages 1–10.
Partridge, D. C., Allman, M., and Floyd, S. (2002). Increasing TCP’s Initial Window. RFC 3390.
Sarpkaya, F. B., Fund, F., and Panwar, S. (2025). To adopt or not to adopt l4s-compatible congestion control? understanding performance in a partial l4s deployment. In Testart, C., van Rijswijk-Deij, R., and Stiller, B., editors, Passive and Active Measurement, pages 217–246, Cham. Springer Nature Switzerland.
Schepper, K. D., Albisser, O., Tilmans, O., and Briscoe, B. (2022). Dual queue coupled aqm: Deployable very low queuing delay for all. Accessed in december 2025.
Schepper, K. D. and Briscoe, B. (2023). The Explicit Congestion Notification (ECN) Protocol for Low Latency, Low Loss, and Scalable Throughput (L4S). RFC 9331.
Schepper, K. D., Briscoe, B., and White, G. (2023). Dual-Queue Coupled Active Queue Management (AQM) for Low Latency, Low Loss, and Scalable Throughput (L4S). RFC 9332.
Shirmarz, A., Bragatto, M. N., Verdi, F. L., Singh, S. K., Rothenberg, C. E., Patra, G., and Pongrácz, G. (2025). In-network ar/cg traffic classification entirely deployed in the programmable data plane: Unlocking rtp features and l4s integration. In 2025 IEEE 11th International Conference on Network Softwarization (NetSoft), pages 477–485.
SpeedTest (2025). Speedtest global index. Internet. Accessed: November 2025.
Team, L. (2025). Kernel tree containing patches for tcp prague and the dualpi2 qdisc. GitHub Repository.
White, G. (2025). Operational Guidance on Coexistence with Classic ECN during L4S Deployment. Internet-Draft draft-ietf-tsvwg-l4sops-08, Internet Engineering Task Force. Work in Progress.
