User Fair Active Queue Management

XU Jian; LI Shan-ping

您当前的位置：

首页 >

文章列表页 >

User Fair Active Queue Management

更新时间：2025-07-16

- User Fair Active Queue Management
- Acta Electronica Sinica Vol. 32, Issue 3, Pages: 435-440(2004)
- 作者机构：
  
  浙江大学计算机学院,浙江,杭州,310027
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP393
- Published：2004
- 稿件说明：
移动端阅览
XU Jian, LI Shan-ping. User Fair Active Queue Management[J]. Acta Electronica Sinica, 2004, 32(3): 435-440.
DOI：

XU Jian, LI Shan-ping. User Fair Active Queue Management[J]. Acta Electronica Sinica, 2004, 32(3): 435-440. DOI：

摘要

用户公平活动队列管理算法UFQ(User Fair Queuing)的目标是在各种网络环境中都能为所有的用户提供满意度一致的服务.UFQ采用在网络边缘标记用户所属数据报的期望服务满意度

在网络核心根据数据报的满意度高低

结合当前数据报流经节点的拥塞程度

来决定数据报的丢弃或标记(使用ECN)

从而获得不同用户一致满意的服务.UFQ不要求接纳控制和信令.它仅在网络边缘保持数据流的状态信息;只维护一个先进先出队列

通过拥塞时丢弃或标记较高满意度的数据报

在不同的用户之间公平地分配网络带宽

从而有效地控制、减轻拥塞.通过TCP/IP网络的模拟

证实了算法能够按照用户期望满意度公平地分配网络带宽

提高网络的服务质量.

Abstract

A user fair active queue management algorithm named UFQ is introduced.UFQ aims at fairly distributing the network resources among users in various situations.Fairness here means different users who have different profiles will get network resources proportionally.User's utility is defined.User's Packets will be inserted a label with its utility in network edge.During network congestion

UFQ will drop or mark packets according to utility inserted in packets.So it distributes network resources among users with average utility.UFQ requires neither admission control nor signaling.Edge router maintains per flow state and core router maintains no flow state.Only one FIFO queue is kept in each node.With simulation and experiments of TCP and UDP traffic

UFQ has shown good performance on quality of service.

关键词

Keywords

references

Views

874

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Fairness in Trustworthy Federated Learning: A Survey

Fair Computation Offloading for a Multi-Antenna NOMA Aided Mobile Edge Computing Network

Fair Provable Data Possession Scheme Based on Blockchain

Self-Supervised Learning for Alleviating Popularity Bias in Recommender Systems

A Survey of Scheduling Mechanisms for Programmable Data Plane in Data Center

Related Author

CHEN Hao-yu

LI Yi-dong

ZHANG Hong-lei

CHEN Nai-yue

ZENG Sheng

HUANG Xiao-hong

LI Dan-dan

YU Wei-jun

Related Institution

Key Laboratory of Big Data & Artificial Intelligence in Transportation， Ministry of Education

School of Computer and Information Technology，Beijing Jiaotong University

Office of Party and Government Affairs， Torch Hi-tech Industrial Development Zone

School of Computer Science（National Pilot Software Engineering School）， Beijing University of Posts and Telecommunications

Department of Cyber Security and Information Law, Chongqing University of Posts and Telecommunications

⁰