Share Email Print
cover

Proceedings Paper

A class-based scheduling algorithm with high throughput for optical burst switching networks
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Optical burst switching (OBS) is more efficient and feasible solution to build terabit IP-over-WDM optical network by employing relatively matured photonic and opto-electronic devices and combining the advantage of high bandwidth of optical transmission/switching and high flexibility of electronic control/processing. Channel scheduling algorithm is one of the key issues related to OBS networks. In this paper, a class-based scheduling algorithm is presented with emphasis on fairly utilizing the bandwidth among different services. A maximum reserved channel numbers and a maximum channel search times is introduced for each service based on its class of services, load and available bandwidth resource in the class-based scheduling algorithm. The performance of the scheduling algorithm is studied in detail by simulation. The results show that the scheduling algorithm can allocate the bandwidth more fairly among different services and the total burst loss ratio under high throughput can be lowered with acceptable expense on delay performance of services with lower delay requirement. Problems related with burst loss ratio and the delay requirement of different services can be well solved simultaneously.

Paper Details

Date Published: 11 February 2005
PDF: 6 pages
Proc. SPIE 5625, Optical Transmission, Switching, and Subsystems II, (11 February 2005); doi: 10.1117/12.570055
Show Author Affiliations
Guiling Wu, Shanghai Jiaotong Univ. (China)
Jianping Chen, Shanghai Jiaotong Univ. (China)
Xinwan Li, Shanghai Jiaotong Univ. (China)
Hui Wang, Shanghai Jiaotong Univ. (China)


Published in SPIE Proceedings Vol. 5625:
Optical Transmission, Switching, and Subsystems II
Cedric F. Lam; Wanyi Gu; Norbert Hanik; Kimio Oguchi, Editor(s)

© SPIE. Terms of Use
Back to Top