A Burst Assembly Method for Decreasing End-to-end Packet Delays in Optical Burst Switching Networks

Abstract

Although the offset-time-based scheme can differentiate packet/burst loss rate according to their QoS levels in optical burst switching networks, it increases end-to-end packet delays of packets in a burst with higher QoS level. This is because it increases offset time of bursts in proportional to their QoS levels. In this paper, we propose a burst assembly method that decreases end-to-end packet delays. Our proposed method includes packets that arrive during the offset time into the currently assembled bursts while the conventional method includes them into the burst assembled next time. Simulation results show that our method offers more than 21% delay reduction for bursts with the highest QoS level.

A model of optimal burst assembly for delay reduction at ingress OBS nodes

“…In his model, Fukushima [9] allows aggregating the packets that arrive during the offset time into the current burst (Figure 1c), and suggests an estimation formula based on the average rate (λ avg ) of M latestarriving packets, as follows:…”

“…The basic models of burst assembly, namely the timer-based model [2], the length-based model [3], and the hybrid model [4,5], do not study reducing the buffering delay; however, the models of burst assembly for delay reduction [6][7][8][9][10][11] try to eliminate the offset time by sending the BCP early, before the completion of its * Correspondence: vvmnhat@hueuni.edu.vn burst. However, it needs to estimate the length of the completed burst, because this information must be carried in the BCP.…”

“…However, it needs to estimate the length of the completed burst, because this information must be carried in the BCP. Various estimation methods have been used in [6][7][8][9][10][11]. However, these models of burst assembly have the following limitations: (1) they always have a certain estimation error that has not yet been minimized; (2) the estimation error of the current assembly has not been used effectively for the next assemblies; (3) the chosen thresholds have a significant impact on the estimation error, yet they have not been studied yet.…”