A typical user is concerned only with the quality of service of a network on an end-to-end basis. Therefore, how end-to-end requirements are mapped into the local switching node requirements and maximum network utilization is a function of network internal design. In this paper, we address the problem of QOS allocation. We derived an optimal QOS allocation policy and decided the maximum utilization bound in a deterministic traffic model. We adopted the worst case delay bound as the end-to-end and local QOS requirement. With (σ, ρ) traffic model, we derived a formula for delay bound and the number of connections. We found that with the delay bound as the QOS metric, there is a significant difference in the performance of allocation policies. We also developed an evaluation strategy to analyze allocation policies. The numerical results for two simple network topologies: tandem network model and uneven traffic load model, compare the equal allocation policy with the optimal allocation policy and show the correctness and efficiency of QOS allocation policy.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Yen-Ping CHU, E-Hong HWANG, Kuan-Cheng LIN, Chin-Hsing CHEN, "Local Allocation of End-to-End Delay Requirement" in IEICE TRANSACTIONS on Communications,
vol. E82-B, no. 9, pp. 1380-1387, September 1999, doi: .
Abstract: A typical user is concerned only with the quality of service of a network on an end-to-end basis. Therefore, how end-to-end requirements are mapped into the local switching node requirements and maximum network utilization is a function of network internal design. In this paper, we address the problem of QOS allocation. We derived an optimal QOS allocation policy and decided the maximum utilization bound in a deterministic traffic model. We adopted the worst case delay bound as the end-to-end and local QOS requirement. With (σ, ρ) traffic model, we derived a formula for delay bound and the number of connections. We found that with the delay bound as the QOS metric, there is a significant difference in the performance of allocation policies. We also developed an evaluation strategy to analyze allocation policies. The numerical results for two simple network topologies: tandem network model and uneven traffic load model, compare the equal allocation policy with the optimal allocation policy and show the correctness and efficiency of QOS allocation policy.
URL: https://global.ieice.org/en_transactions/communications/10.1587/e82-b_9_1380/_p
Copy
@ARTICLE{e82-b_9_1380,
author={Yen-Ping CHU, E-Hong HWANG, Kuan-Cheng LIN, Chin-Hsing CHEN, },
journal={IEICE TRANSACTIONS on Communications},
title={Local Allocation of End-to-End Delay Requirement},
year={1999},
volume={E82-B},
number={9},
pages={1380-1387},
abstract={A typical user is concerned only with the quality of service of a network on an end-to-end basis. Therefore, how end-to-end requirements are mapped into the local switching node requirements and maximum network utilization is a function of network internal design. In this paper, we address the problem of QOS allocation. We derived an optimal QOS allocation policy and decided the maximum utilization bound in a deterministic traffic model. We adopted the worst case delay bound as the end-to-end and local QOS requirement. With (σ, ρ) traffic model, we derived a formula for delay bound and the number of connections. We found that with the delay bound as the QOS metric, there is a significant difference in the performance of allocation policies. We also developed an evaluation strategy to analyze allocation policies. The numerical results for two simple network topologies: tandem network model and uneven traffic load model, compare the equal allocation policy with the optimal allocation policy and show the correctness and efficiency of QOS allocation policy.},
keywords={},
doi={},
ISSN={},
month={September},}
Copy
TY - JOUR
TI - Local Allocation of End-to-End Delay Requirement
T2 - IEICE TRANSACTIONS on Communications
SP - 1380
EP - 1387
AU - Yen-Ping CHU
AU - E-Hong HWANG
AU - Kuan-Cheng LIN
AU - Chin-Hsing CHEN
PY - 1999
DO -
JO - IEICE TRANSACTIONS on Communications
SN -
VL - E82-B
IS - 9
JA - IEICE TRANSACTIONS on Communications
Y1 - September 1999
AB - A typical user is concerned only with the quality of service of a network on an end-to-end basis. Therefore, how end-to-end requirements are mapped into the local switching node requirements and maximum network utilization is a function of network internal design. In this paper, we address the problem of QOS allocation. We derived an optimal QOS allocation policy and decided the maximum utilization bound in a deterministic traffic model. We adopted the worst case delay bound as the end-to-end and local QOS requirement. With (σ, ρ) traffic model, we derived a formula for delay bound and the number of connections. We found that with the delay bound as the QOS metric, there is a significant difference in the performance of allocation policies. We also developed an evaluation strategy to analyze allocation policies. The numerical results for two simple network topologies: tandem network model and uneven traffic load model, compare the equal allocation policy with the optimal allocation policy and show the correctness and efficiency of QOS allocation policy.
ER -