QRED: an enhancement approach for congestion control in network communications
- PDF / 940,457 Bytes
- 7 Pages / 595.276 x 790.866 pts Page_size
- 103 Downloads / 159 Views
ORIGINAL RESEARCH
QRED: an enhancement approach for congestion control in network communications Dharamdas Kumhar1
•
Avanish kumar2 • Anil Kewat1
Received: 8 November 2019 / Accepted: 3 October 2020 Bharati Vidyapeeth’s Institute of Computer Applications and Management 2020
Abstract This Nowadays, along with the rapid development of Internet technology, the user’s network functionality and need for internet performance are also growing very fast. Therefore network congestion being increasingly serious and congestion control is becoming a more important issue. To detect and control congestion many queue management techniques are accessible; the extremely fundamental of them is Random Early Detection (RED). In order to enhance network performance, Quadratic Random Early Detection (QRED) is proposed. This paper gives comparative analysis between RED and QRED and shows that proposed method has improved network performance in the congested network. Keywords Congestion control Random early detection Quadratic random early detection Throughput Packet loss Jitter
& Dharamdas Kumhar [email protected] Avanish kumar [email protected] Anil Kewat [email protected] 1
Department of Mathematical Sciences and Computer Applications, Bundelkhand University, Jhansi, U.P., India
2
Commission for Scientific and Technical Terminology, Ministry of Human Resource Development, RK Puram, New Delhi, India
1 Introduction In data transmission speed is a very important issue nowadays. Speed is increasing on the demands of highspeed Internet applications to transmit huge data through the network. To transfer data from one side to the other, network produces mandatory extremely high-speed router buffers. Therefore router buffers work on forwarding data from one network to another on the path from a source application to the target [1]. In router buffers, congestion is one of the essential problems that may arise because of arriving packets which are not homogeneous by the available network resources i.e. bandwidth [2]. Network performance decay due to congestion by having a low throughput, maximum packets lost, maximum mean waiting time and instability of queue length at the router buffers. It is hereby explicit that in consecutively to enhance network performance; various methods essential be devised to intercept the network. The brilliance congestion control mechanism is Drop Tail. Drop Tail is used for a small portion of the bandwidth. The limitations of using Drop Tail are full queue and lockout. The lockout drawback may occur when Drop Tail allows a few connections to dominate queue space. Full queue event may occur when there will be high queuing delays at a congested router. Since, when the router buffer has become full they send back traffic jam indication to sources by means of packets losses. The full queue mechanism may maintain for a long time under Drop Tail queue mechanism. Hereby, it is essential for a router to preserve a small queue length i.e. bandwidth. Therewith, it is essential to dete
Data Loading...
