Scheduling and resources allocation in network traffic using multiobjective, multiuser joint traffic engineering

  • PDF / 2,049,230 Bytes
  • 13 Pages / 595.276 x 790.866 pts Page_size
  • 21 Downloads / 231 Views

DOWNLOAD

REPORT


(0123456789().,-volV) (0123456789().,-volV)

Scheduling and resources allocation in network traffic using multiobjective, multiuser joint traffic engineering Ghazanfar Latif1 • N. Saravanakumar2 • Jaafar Alghazo1 • P. Bhuvaneswari3 • K. Shankar4 Muhammad O. Butt5



 Springer Science+Business Media, LLC, part of Springer Nature 2020

Abstract Reciprocal optimality is a desirable characteristic from the end user perspective, for both Best Effort and Quality of Service (QoS) datagrams delivery within a network. A cohesive solution to such issues is the expected outcome of this work. It is also aimed to formulate and evaluate enhanced scheduling algorithm for packets delivery, focusing on maximal user satisfaction and minimal networking operation costs. Mathematical optimization techniques have been applied to Application, Transport and Network layers of Open System Interconnection model. Novel elucidation of optimal packets switching, traffic engineering, congestion avoidance and QoS improvements are discussed. A simplified routing mechanism is devised, along with the formulation and analysis of nonlinear constrained Multiobjective, multiuser joint traffic engineering case. This approach would be useful for networking and digital communication domains. Keywords Reciprocal optimality  Congestion avoidance  Quality of service (QoS)  Optimal packets switching

1 Introduction A typical information quantity (sequence of bits), is usually referred to as packet or datagram. The packets undergo delays while travelling from source to destination due to throughput perturbations (propagation delay), networking devices processing (queueing delay), and wave propagation through network links. Such networks can be easily modeled as a graph. From the end user perspective, the traffic

& Jaafar Alghazo [email protected] 1

College of Computer Engineering and Sciences, Prince Mohammad Bin Fahd University, Khobar, Saudi Arabia

2

Department of ECE, Dr Mahalingam College of Engineering and Technology, Pollachi, India

3

Department of ECE, Sri Venkateswara College of Engineering and Technology (Autonomous), Chittoor, India

4

Department of ECE, Annamacharya Institute of Technology and Sciences, Rajampet, India

5

College of Engineering, Prince Mohammad Bin Fahd University, Khobar, Saudi Arabia

stream must satisfy certain application timing requirements. However, for the network operator there is certain cost of effort to provide a particular throughput (bit rate) to each user on a link, keeping in view the overall capacity of the link. Bandwidth is regarded as the maximum possible bits transfer rate for a particular single path (link), measured as bits/s. However, throughput is the actual bit transfer rate on a particular link, also measured as bits/s. Scheduling can be regarded as the allocation of a particular link to a particular user. Latency/delay is measured in seconds, and packets length is measured in bits. Congestion in a computer network occurs when there is a mismatch between particular links transmission rates. Pack