Pipelined circuit switching: Analysis for the torus with non-uniform traffic

2008 ◽  
Vol 54 (1-2) ◽  
pp. 97-110 ◽  
Author(s):  
F. Safaei ◽  
A. Khonsari ◽  
M. Fathy ◽  
M. Ould-Khaoua
2001 ◽  
Vol 02 (04) ◽  
pp. 471-483 ◽  
Author(s):  
GEYONG MIN ◽  
MOHAMED OULD-KHAOUA

Performance properties of switching methods used in existing interconnection networks have been mainly examined under the assumption that network traffic follows the Poisson arrival process. However, the Poisson process is unable to capture the bursty nature of traffic. In this paper we propose the first analytical model for pipelined circuit switching under bursty traffic using the Markov-Modulated Poisson Process (MMPP). Simulation experiments reveal that the proposed model exhibits a good degree of accuracy for various networks sizes and under different traffic conditions.


2003 ◽  
Vol 04 (02) ◽  
pp. 179-197 ◽  
Author(s):  
Geyong Min ◽  
Mohamed Ould-Khaoua ◽  
Lewis M. Mackenzie

Switching techniques have a strong effect on the performance of interconnection networks. Many studies have shown that traffic loads in parallel computation environments can exhibit burstiness property. In order to obtain a deep understanding of the efficiency of switching techniques, this study proposes an analytical performance model for pipelined circuit switching (PCS) in binary n-cube networks under bursty traffic, which is modeled by a switched Poisson process (SPP). Simulation experiments demonstrate that the proposed model exhibits a good degree of accuracy. This model and the one proposed in Ref. [13] for wormhole switching (WS) are then applied to investigate the relative performance merits of PCS and WS in the presence of bursty loads.


Sign in / Sign up

Export Citation Format

Share Document