The Last WAVE: Integrating with Mininet for More Flexible and Reproducible Network Experiments
Resumo
Experimentation is essential in computer network research because it enables controlled and reproducible evaluation of applications and protocols. This paper presents the third version of WAVE (Workload Assay for Verified Experiments), evolving from the original multiple-workload generator and its later extensions with new load models and microburst support. WAVE controls the execution of real application instances over time according to mathematical workload models, currently supporting sinusoid, flashcrowd, and step patterns, as well as video-oriented workloads and microbursts. The main contribution of this version is the native integration with Mininet, which removes the previous integration barrier with simulated environments and allows workload generation and network emulation to be configured in a unified workflow. In this integrated setting, researchers can define different topologies, such as linear and tree, and inject realistic network parameters, including additional delay and packet loss. These capabilities provide a more flexible and reproducible environment for evaluating distributed applications and network behavior under diverse experimental conditions.
Referências
Beuttenmuller, D. C., Valério, M. F. d. A., Silva, C. L. L. T., Da Silva, I. M., Maciel Jr., P. D., and De Almeida, L. C. (2025). A new wave: Exploring new load pattern models for experimentation in computer networks. In Anais do Salão de Ferramentas do Simpósio Brasileiro de Redes de Computadores e Sistemas Distribuídos (SBRC), pages 11–22, Natal/RN. Sociedade Brasileira de Computação. ISSN 2177-9384.
Hafez, N. A., Hassan, M. S., and Landolsi, T. (2023). Reinforcement learning-based rate adaptation in dynamic video streaming. Telecommunication Systems, 83(4):395–407.
Kim, M. and Chung, K. (2022). Reinforcement Learning-Based adaptive streaming scheme with edge computing assistance. Sensors (Basel), 22(6).
Lantz, B., Heller, B., and McKeown, N. (2010). Mininet: An instant virtual network on your laptop (or other pc). In Proceedings of the 9th ACM SIGCOMM Workshop on Hot Topics in Networks, pages 1–6.
Lin, H., Shen, Z., Zhou, H., Liu, X., Zhang, L., Xiao, G., and Cheng, Z. (2020). Knn-q learning algorithm of bitrate adaptation for video streaming over http. In 2020 Information Communication Technologies Conference (ICTC), pages 302–306.
Sandvine (2023). Global Internet Phenomena. Technical report, Sandvine.
Spang, B. et al. (2023). Sammy: Smoothing video traffic to be a friendly internet neighbor. In Proceedings of the ACM SIGCOMM 2023 Conference, page 754–768, New York, NY, USA. Association for Computing Machinery.
Wei, X. et al. (2021). Reinforcement learning-based QoE-oriented dynamic adaptive streaming framework. Information Sciences, 569:786–803.
