ANALYTICAL MODELS BASED DISCRETE-TIME QUEUEING FOR THE CONGESTED NETWORK

Author:

AL-DIABAT MOFLEH1,ABDEL-JABER HUSSEIN2,THABTAH FADI3,ABOU-RABIA OSMAN4,KISHTA MAHMOUD5

Affiliation:

1. Computer Science Department, Al-al-Bayt University, Jordan

2. Department of Computer Science, The World Islamic Sciences & Education (W.I.S.E.) University, Amman, Jordan

3. Management Information System Department, Philadelphia University, Amman, Jordan

4. Department of Mathematics and Computer Science, Laurentian University, Sudbury ON, Canada

5. Dean of Academic Research, Philadelphia University, Amman, Jordan

Abstract

Congestion is one of the well-studied problems in computer networks, which occurs when the request for network resources exceeds the buffer capacity. Many active queue management techniques such as BLUE and RED have been proposed in the literature to control congestions in early stages. In this paper, we propose two discrete-time queueing network analytical models to drop the arrival packets in preliminary stages when the network becomes congested. The first model is based on Lambda Decreasing and it drops packets from a probability value to another higher value according to the buffer length. Whereas the second proposed model drops packets linearly based on the current queue length. We compare the performance of both our models with the original BLUE in order to decide which of these methods offers better quality of service. The comparison is done in terms of packet dropping probability, average queue length, throughput ratio, average queueing delay, and packet loss rate.

Publisher

World Scientific Pub Co Pte Lt

Subject

Computer Science Applications,Modeling and Simulation

Reference9 articles.

1. M. Welzl, Network Congestion Control: Managing Internet Traffic (John Wiley & Sons, 1995) p. 282.

2. The BLUE active queue management algorithms

3. A control theoretic approach to active queue management

4. Random early detection gateways for congestion avoidance

Cited by 5 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Linear Random Early Detection for Congestion Control at the Router Buffer;Informatica;2022-04-19

2. An Exponential Active Queue Management Method Based on Random Early Detection;Journal of Computer Networks and Communications;2020-05-22

3. Multi-Criteria Evaluation and Benchmarking for Active Queue Management Methods: Open Issues, Challenges and Recommended Pathway Solutions;International Journal of Information Technology & Decision Making;2019-07

4. Performance study of Active Queue Management methods: Adaptive GRED, REDD, and GRED-Linear analytical model;Journal of King Saud University - Computer and Information Sciences;2015-10

5. Gentle-BLUE: A New Method for Active Queue Management;2014 3rd International Conference on Advanced Computer Science Applications and Technologies;2014-12

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3