Erasure Coding-Based Non-Conservative Network Communication: A Ground Up Approach
Xuhong Cai, Yi Chen, Shenghao Yang, Xingyan Shi
Abstract
Erasure coding has been studied for network communications due to various advantages over retransmission. Many prior works have discussed how to integrate erasure coding into existing communication protocols. However, as the congestion control mechanisms inherent to these established protocols were not natively designed for coded flows, the potential of erasure coding may not be fully realized. In this paper, we demonstrate the benefits of incorporating erasure coding into congestion control design. We explore a communication protocol that is built from the ground up with erasure coding, named Coding-based Non-conservative Communication Protocol (CNCP). CNCP adopts hop-by-hop congestion control and distinguishes itself from traditional conservative flow enforcement by accommodating non-conservative flows. To address potential issues associated with non-conservative flows, such as increased congestion and queue instability, CN CP utilizes a specific queuing design and packet deletion strategy at intermediate network nodes. Our analysis reveals that non-conservative flows do not undermine fairness and may expedite convergence towards optimal network utilization. Furthermore, we develop a hop-by-hop congestion control mechanism based on primal-dual method, specifically tailored for non-conservative flows in CNCP. Through simulations, we show that combining erasure coding and non-conservative flow control can significantly enhance network communication performance under dynamic network conditions.
Ask about this paper
Ask your agent about it.
Lune has read the top-tier papers around this one, so every answer names the papers it rests on.
Your agent calls
Lunesearch_papers
Free to start. No credit card required.
Terminal
Install the CLIlune papers get 417f536e-8982-4754-8bac-ca357d5ebb51Related papers
- ChameleonEC: Exploiting Tunability of Erasure Coding for Low-Interference RepairYuhui Cai, Shiyao Lin, Zhirong Shen, Jiahui Yang et al.HPCA 2025 · 5 citations
- CRaft: An Erasure-coding-supported Version of Raft for Reducing Storage Cost and Network CostZizhong Wang, Tongliang Li, Haixia Wang, Airan Shao et al.FAST 2020 · 42 citations
- Uno: A One-Stop Solution for Inter- and Intra-Data Center Congestion Control and Reliable ConnectivityTommaso Bonato, Sepehr Abdous, Abdul Kabbani, Ahmad Ghalayini et al.SC 2025 · 7 citations
- Enhanced Flooding-Based Routing Protocol for Swarm UAV Networks: Random Network Coding Meets ClusteringHao Song, Lingjia Liu, Bodong Shang, Scott Pudlewski et al.INFOCOM 2021 · 32 citations
- CodedBulk: Inter-Datacenter Bulk Transfers using Network CodingShih-Hao Tseng, Saksham Agarwal, Rachit Agarwal, Hitesh Ballani et al.NSDI 2021 · 18 citations
