Impact of burst control packet congestion on burst loss rate in optical burst switched networks

Authors:
In-Yong Hwang;Seoungyoung Lee;Hong-Shik Park
Affiliations:
School of Engineering, Information and Communications University, Daejeon, Korea;School of Engineering, Information and Communications University, Daejeon, Korea;School of Engineering, Information and Communications University, Daejeon, Korea
Venue:
ICOIN'06 Proceedings of the 2006 international conference on Information Networking: advances in Data Communications and Wireless Networks
Year:
2006

Citing 6
Cited 1

Optical burst switching (OBS) - a new paradigm for an optical Internet

Journal of High Speed Networks - Special issue on optical networking
Delay bounds for a network of guaranteed rate servers with FIFO aggregation

Computer Networks: The International Journal of Computer and Telecommunications Networking
Delay Bounds in a Network with Aggregate Scheduling

QofIS '00 Proceedings of the First COST 263 International Workshop on Quality of Future Internet Services
Fundamental Trade-offs in Aggregate Packet Scheduling

ICNP '01 Proceedings of the Ninth International Conference on Network Protocols
Control architecture in optical burst-switched WDM networks

IEEE Journal on Selected Areas in Communications
Optical burst switching: a new area in optical networking research

IEEE Network: The Magazine of Global Internetworking

A performance comparison of OPS with variable-size packets and OBS

Optical Switching and Networking

Quantified Score

Hi-index	0.00

Visualization

Abstract

In the Optical Burst Switching (OBS) research area, the burst control packet (BCP) queuing delay problem has not been addressed because it is believed to be quite small However, with a realistic OBS simulator, we have investigated this issue and clarified its impact on performance degradation from the point-of-view of data burst loss rate We know that the BCP load to the control channel is not negligible Thus, the burst loss rate due to the queueing delay of BCP on the control channel is very serious compared to the existing well-known burst contention We propose a Dynamic Offset-Time Update scheme to avoid serious data burst loss due to BCP queueing delay, and a Priority BCP queue to guarantee the minimum offset-time for high class bursts Our simulation results show that the Dynamic Offset-Time Update scheme can completely avoid data burst loss due to BCP congestion while guaranteeing a certain level of minimum offset-time.