Discrete and Continuous Models and Applied Computational ScienceDiscrete and Continuous Models and Applied Computational ScienceDiscrete and Continuous Models and Applied Computational Science2658-46702658-7149Peoples' Friendship University of Russia named after Patrice Lumumba (RUDN University)5292510.22363/2658-4670-2026-34-3-379-395RUDFRIModeling and SimulationМатематическое моделированиеResearch ArticleComparative analysis of active queue management algorithms: RED, ARED, Gentle REDСравнительный анализ алгоритмов активного управления очередями: RED, ARED, Gentle REDhttps://orcid.org/0009-0007-7009-7038PalagashviliAbuli M.ПалагашвилиА. М.

PhD Student of Department of Department of Probability Theory and Cyber Security of RUDN University

1042250185@rudn.ruhttps://orcid.org/0000-0001-6107-6461ErmolayevaAnna M.ЕрмолаеваА. М.

Assistent Professor of Department of Probability Theory and Cyber Security of RUDN University

ermolaeva-am@rudn.ruhttps://orcid.org/0000-0003-4466-8531VelievaTatyana R.ВелиеваТ. Р.

Candidate of Physical and Mathematical Sciences, Associate Professor of Department of Probability Theory and Cyber Security of RUDN University

velieva-tr@rudn.ruRUDN UniversityРоссийский университет дружбы народов15102026343VOL 34, NO3 (2026)ТОМ 34, №3 (2026)37939519102026Copyright ©; 2026, Palagashvili A.M., Ermolayeva A.M., Velieva T.R.Copyright ©; 2026, Палагашвили А.М., Ермолаева А.М., Велиева Т.Р.2026Palagashvili A.M., Ermolayeva A.M., Velieva T.R.Палагашвили А.М., Ермолаева А.М., Велиева Т.Р.https://creativecommons.org/licenses/by-nc/4.0https://journals.rudn.ru/miph/article/view/52925

Background. Active Queue Management (AQM) algorithms of the RED family remain widely deployed in network equipment, yet a systematic three-way comparison of RED, Adaptive RED (ARED), and Gentle RED under strictly identical conditions is lacking in the literature. Purpose. The aim of this study is to quantify the differences in buffer management dynamics, throughput stability, and congestion signaling among the three algorithms and to formulate practical selection guidelines. Method. A dumbbell-topology network model was implemented in the ns-3 discrete-event simulator (version 3.41) with ten TCP bulk-send sources creating sustained congestion at a 2 Mbit/s bottleneck link. All three algorithms were evaluated with identical queue thresholds, load profile, and topology over a 60-second simulation. An extended metric set including EWMA zone distribution and temporal drop structure was applied. Results. All algorithms achieved approximately 90% link utilization (mean throughput 1.80 Mbit/s). RED provided the most predictable throughput (coefficient of variation 18.7%) at the cost of the highest drop count (1240 packets). ARED minimized drops (1003 packets, 19% fewer than RED) but produced the largest queue oscillations (p99 of 49 packets). Gentle RED achieved the lowest mean queue length (6.40 packets) with the highest throughput variability (coefficient of variation 24.7%). Conclusions. No single algorithm is universally optimal; the choice depends on scenario priorities: Gentle RED for minimum delay, ARED for minimum packet loss, and RED for maximum throughput predictability.

Предпосылки. Алгоритмы активного управления очередями (AQM) семейства RED широко применяются в сетевом оборудовании, однако систематическое трёхстороннее сравнение RED, Adaptive RED (ARED) и Gentle RED в строго идентичных условиях недостаточно представлено в литературе. Цель. Целью данного исследования является количественное определение различий в динамике управления буфером, стабильности пропускной способности и характере сигналов перегрузки между тремя алгоритмами, а также формулировка практических рекомендаций по их выбору. Метод. Реализована модель сети с топологией «гантель» в среде дискретно-событийного симулятора ns-3 (версия 3.41) с десятью TCP-источниками, создающими устойчивую перегрузку узкого канала (2 Мбит/с). Все три алгоритма оценены с идентичными порогами очереди, профилем нагрузки и топологией в течение 60 секунд моделирования. Применён расширенный набор метрик, включающий зональное распределение EWMA-средней и временну́ю структуру сбросов. Результаты. Все алгоритмы обеспечили приблизительно 90% утилизации канала (средняя пропускная способность 1,80 Мбит/с). RED показал наиболее предсказуемую пропускную способность (коэффициент вариации 18,7%) ценой наибольшего числа сбросов (1240 пакетов). ARED минимизировал сбросы (1003 пакета, на 19% меньше RED), но породил наибольшие осцилляции очереди (p99 = 49 пакетов). Gentle RED достиг наименьшей средней длины очереди (6,40 пакета) при наибольшей вариабельности пропускной способности (коэффициент вариации 24,7%). Выводы. Универсально оптимального алгоритма не существует; выбор определяется приоритетами сценария: Gentle RED для минимальной задержки, ARED для минимальных потерь пакетов, RED для максимальной предсказуемости пропускной способности.

active queue managementAQMREDAdaptive REDGentle REDcongestion controlns-3network simulationbufferbloatактивное управление очередямиAQMREDAdaptive REDGentle REDуправление перегрузкойns-3сетевое моделированиеbufferbloatJ. Gettys and K. Nichols, “Bufferbloat: Dark Buffers in the Internet,” Communications of the ACM, vol. 55, no. 1, pp. 57-65, 2012. DOI: 10.1145/2063176.2063196K. Nichols, V. Jacobson, A. McGregor, and J. Iyengar, Controlled Delay Active Queue Management, RFC 8289, 2018.B. Braden, D. Clark, J. Crowcroft, et al., Recommendations on Queue Management and Congestion Avoidance in the Internet, RFC 2309, 1998.R. Adams, “Active Queue Management: A Survey,” IEEE Communications Surveys & Tutorials, vol. 15, no. 3, pp. 1425-1476, 2013. DOI: 10.1109/SURV.2012.082212.00018A. V. Korolkova, D. S. Kulyabov, and T. I. Tchernyshev, “On the Classification of RED Algorithm Modifications,” Bulletin of Peoples’ Friendship University of Russia. Series: Mathematics, Informatics, Physics, vol. 26, no. 3, pp. 305-316, 2018.K. Nichols and V. Jacobson, “Controlling Queue Delay,” ACM Queue, vol. 10, no. 5, pp. 20-34, 2012. DOI: 10.1145/2209249.2209264R. Pan, P. Natarajan, C. Piglione, M. S. Prabhu, V. Subber, F. Khan, and K. Athiappan, “PIE: A Lightweight Control Scheme to Address the Bufferbloat Problem,” in Proceedings of IEEE International Conference on High Performance Switching and Routing (HPSR), 2013, pp. 148-155. DOI: 10.1109/HPSR.2013.6602305S. Floyd and V. Jacobson, “Random Early Detection Gateways for Congestion Avoidance,” IEEE/ACM Transactions on Networking, vol. 1, no. 4, pp. 397-413, 1993. DOI: 10.1109/90.251892V. Jacobson, “Congestion Avoidance and Control,” ACM SIGCOMM Computer Communication Review, vol. 18, no. 4, pp. 314-329, 1988. DOI: 10.1145/52325.52356M. Christiansen, K. Jeffay, D. Ott, and F. D. Smith, “Tuning RED for Web Traffic,” IEEE/ACM Transactions on Networking, vol. 9, no. 3, pp. 249-264, 2001. DOI: 10.1109/90.929849S. Floyd, R. Gummadi, and S. Shenker, “Adaptive RED: An Algorithm for Increasing the Robustness of RED’s Active Queue Management,” ICSI, Tech. Rep., 2001.S. Floyd. “Recommendation on using the “gentle” variant of RED. ”[Online]. Available: https: //www.icir.org/floyd/red/gentle.htmlK. K. Ramakrishnan, S. Floyd, and D. Black, The Addition of Explicit Congestion Notification (ECN) to IP, RFC 3168, 2001.A. V. Korolkova, Mathematical Model of Traffic Transmission with Dynamic Flow Intensity Regulated by a RED-type Algorithm: PhD Thesis. Moscow: RUDN University, 2010.T. R. Velieva, Analysis of Self-Oscillation Modes in Continuous-Discrete Dynamical Models: PhD Thesis. Moscow: RUDN University, 2019.W.-c. Feng, D. D. Kandlur, D. Saha, and K. G. Shin, “BLUE: A New Class of Active Queue Management Algorithms,” University of Michigan, Tech. Rep. CSE-TR-387-99, 2002.T. J. Ott, T. V. Lakshman, and L. H. Wong, “SRED,” in Proceedings of IEEE INFOCOM, vol. 3, 1999, pp. 1346-1355. DOI: 10.1109/INFCOM.1999.752153C. V. Hollot, V. Misra, D. Towsley, and W.-B. Gong, “On Designing Improved Controllers for AQM Routers Supporting TCP Flows,” in Proceedings of IEEE INFOCOM, 2001, pp. 1726-1734. DOI: 10.1109/INFCOM.2001.916670V. Misra, W.-B. Gong, and D. Towsley, “Fluid-based Analysis of a Network of AQM Routers Supporting TCP Flows with an Application to RED,” ACM SIGCOMM Computer Communication Review, vol. 30, no. 4, pp. 151-160, 2000. DOI: 10.1145/347059.347421T. ns-3 Project. “ns-3: A Discrete-Event Network Simulator.” Version 3.41. [Online]. Available: https://www.nsnam.org/A. Varga and R. Hornig, “An Overview of the OMNeT++ Simulation Environment,” in Proceedings of the 1st International Conference on Simulation Tools and Techniques (SIMUTools), 2008, pp. 1-10. DOI: 10.4108/ICST.SIMUTOOLS2008.3027S. Ha, I. Rhee, and L. Xu, “CUBIC: A New TCP-Friendly High-Speed TCP Variant,” in ACM SIGOPS Operating Systems Review, vol. 42, 2008, pp. 64-74. DOI: 10.1145/1400097.1400105R. Jain, “Throughput Fairness Index,” IEEE 802.11-98/S41, 1999.