It is always difficult to monitor stringent cell loss ratio (CLR), e.g. 10-9, in asynchronous transfer mode (ATM) networks, because its measurement period is too long for real-time measurement. In this paper, we propose new performance monitoring mechanisms for stringent CLR. We consider virtual ATM switches whose resources are much smaller than the real system and thus much higher CLRs will be obtained within a relatively short measurement period. By applying some regression methods in the CLRs obtained from the virtual system, we can estimate the actual CLR of the real system quite accurately and our performance monitoring mechanisms will be operated based on the estimation. Through the numerical examples, our mechanisms are not only more accurate than the traditional methods, but also have shorter measurement periods compared with direct measurement.
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
Hon-Wai CHU, Chi-Chung CHEUNG, Danny H. K. TSANG, "Performance Monitoring Using Regression Approach in ATM Networks" in IEICE TRANSACTIONS on Communications,
vol. E84-B, no. 2, pp. 294-304, February 2001, doi: .
Abstract: It is always difficult to monitor stringent cell loss ratio (CLR), e.g. 10-9, in asynchronous transfer mode (ATM) networks, because its measurement period is too long for real-time measurement. In this paper, we propose new performance monitoring mechanisms for stringent CLR. We consider virtual ATM switches whose resources are much smaller than the real system and thus much higher CLRs will be obtained within a relatively short measurement period. By applying some regression methods in the CLRs obtained from the virtual system, we can estimate the actual CLR of the real system quite accurately and our performance monitoring mechanisms will be operated based on the estimation. Through the numerical examples, our mechanisms are not only more accurate than the traditional methods, but also have shorter measurement periods compared with direct measurement.
URL: https://global.ieice.org/en_transactions/communications/10.1587/e84-b_2_294/_p
Copy
@ARTICLE{e84-b_2_294,
author={Hon-Wai CHU, Chi-Chung CHEUNG, Danny H. K. TSANG, },
journal={IEICE TRANSACTIONS on Communications},
title={Performance Monitoring Using Regression Approach in ATM Networks},
year={2001},
volume={E84-B},
number={2},
pages={294-304},
abstract={It is always difficult to monitor stringent cell loss ratio (CLR), e.g. 10-9, in asynchronous transfer mode (ATM) networks, because its measurement period is too long for real-time measurement. In this paper, we propose new performance monitoring mechanisms for stringent CLR. We consider virtual ATM switches whose resources are much smaller than the real system and thus much higher CLRs will be obtained within a relatively short measurement period. By applying some regression methods in the CLRs obtained from the virtual system, we can estimate the actual CLR of the real system quite accurately and our performance monitoring mechanisms will be operated based on the estimation. Through the numerical examples, our mechanisms are not only more accurate than the traditional methods, but also have shorter measurement periods compared with direct measurement.},
keywords={},
doi={},
ISSN={},
month={February},}
Copy
TY - JOUR
TI - Performance Monitoring Using Regression Approach in ATM Networks
T2 - IEICE TRANSACTIONS on Communications
SP - 294
EP - 304
AU - Hon-Wai CHU
AU - Chi-Chung CHEUNG
AU - Danny H. K. TSANG
PY - 2001
DO -
JO - IEICE TRANSACTIONS on Communications
SN -
VL - E84-B
IS - 2
JA - IEICE TRANSACTIONS on Communications
Y1 - February 2001
AB - It is always difficult to monitor stringent cell loss ratio (CLR), e.g. 10-9, in asynchronous transfer mode (ATM) networks, because its measurement period is too long for real-time measurement. In this paper, we propose new performance monitoring mechanisms for stringent CLR. We consider virtual ATM switches whose resources are much smaller than the real system and thus much higher CLRs will be obtained within a relatively short measurement period. By applying some regression methods in the CLRs obtained from the virtual system, we can estimate the actual CLR of the real system quite accurately and our performance monitoring mechanisms will be operated based on the estimation. Through the numerical examples, our mechanisms are not only more accurate than the traditional methods, but also have shorter measurement periods compared with direct measurement.
ER -