Performance evaluation of prioritized scheduling with buffer management for differentiated services architectures

  • Authors:
  • Ahmed E. Kamal;Hossam S. Hassanein

  • Affiliations:
  • Department of Electrical and Computer Engineering, Iowa State University, 319 Durham, Ames, IA;School of Computing, Queen's University, Kingston, ON, Canada K7L 3N6

  • Venue:
  • Computer Networks: The International Journal of Computer and Telecommunications Networking
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

Differentiated services (DiffServ) is an architecture for the Internet in which various applications are supported using a simple classification scheme. Packets entering the DiffServ domain are marked depending on the packets' class. In this paper we introduce a versatile service and buffering scheme for different classes in a differentiated services Internet. The objective of this scheme is to introduce relative differentiation in terms of mean delay, throughput. In this scheme both high-buffer-priority (HPC) and low-buffer-priority (LPC) classes have reserved buffers. In addition, they have a shared buffer, where the priority of the shared buffer occupancy is to HPC traffic. Service priorities can be for high-buffer-priority traffic, low-buffer-priority traffic or round robin. An exact performance model for the proposed scheme is introduced. The performance model represents HPC and LPC traffic arrivals by a discrete batch Markov arrival process (D-BMAP). The model is used to obtain loss ratios, packet delays and throughputs for both HPC and LPC traffic. Performance results show that implementing buffer priorities, with a weighted round robin packet scheduling policy, results in low packet loss ratio and/or delay for the high-priority class without starving the low-priority class.