We use end-to-end measurements to address the problem of fault diagnosis in computer networks. Since link-level characteristics cannot be uniquely determined from available end-to-end measurements, most existing diagnosis approaches make statistical assumptions of the network to obtain a unique solution. However, the performance of these approaches is not assured due to the uncertainty of the assumptions. Thus the diagnostic accuracy cannot be guaranteed. In this paper, we propose a different paradigm for fault diagnosis which can find all identifiable links and the minimal identifiable link sequences, and infer their loss rates with the least error. Compared with a former representative diagnosis method through experiments, the experimental results show that our method has smaller diagnosis granularity and much less running time for most network topologies. We also conducted experiments using 105 Planetlab hosts. The results validate the performance of our method as well.
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
Yan QIAO, Xuesong QIU, Luoming MENG, "Accurate Diagnosis in Computer Networks Using Unicast End-to-End Measurements" in IEICE TRANSACTIONS on Communications,
vol. E96-B, no. 2, pp. 522-532, February 2013, doi: 10.1587/transcom.E96.B.522.
Abstract: We use end-to-end measurements to address the problem of fault diagnosis in computer networks. Since link-level characteristics cannot be uniquely determined from available end-to-end measurements, most existing diagnosis approaches make statistical assumptions of the network to obtain a unique solution. However, the performance of these approaches is not assured due to the uncertainty of the assumptions. Thus the diagnostic accuracy cannot be guaranteed. In this paper, we propose a different paradigm for fault diagnosis which can find all identifiable links and the minimal identifiable link sequences, and infer their loss rates with the least error. Compared with a former representative diagnosis method through experiments, the experimental results show that our method has smaller diagnosis granularity and much less running time for most network topologies. We also conducted experiments using 105 Planetlab hosts. The results validate the performance of our method as well.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.E96.B.522/_p
Copy
@ARTICLE{e96-b_2_522,
author={Yan QIAO, Xuesong QIU, Luoming MENG, },
journal={IEICE TRANSACTIONS on Communications},
title={Accurate Diagnosis in Computer Networks Using Unicast End-to-End Measurements},
year={2013},
volume={E96-B},
number={2},
pages={522-532},
abstract={We use end-to-end measurements to address the problem of fault diagnosis in computer networks. Since link-level characteristics cannot be uniquely determined from available end-to-end measurements, most existing diagnosis approaches make statistical assumptions of the network to obtain a unique solution. However, the performance of these approaches is not assured due to the uncertainty of the assumptions. Thus the diagnostic accuracy cannot be guaranteed. In this paper, we propose a different paradigm for fault diagnosis which can find all identifiable links and the minimal identifiable link sequences, and infer their loss rates with the least error. Compared with a former representative diagnosis method through experiments, the experimental results show that our method has smaller diagnosis granularity and much less running time for most network topologies. We also conducted experiments using 105 Planetlab hosts. The results validate the performance of our method as well.},
keywords={},
doi={10.1587/transcom.E96.B.522},
ISSN={1745-1345},
month={February},}
Copy
TY - JOUR
TI - Accurate Diagnosis in Computer Networks Using Unicast End-to-End Measurements
T2 - IEICE TRANSACTIONS on Communications
SP - 522
EP - 532
AU - Yan QIAO
AU - Xuesong QIU
AU - Luoming MENG
PY - 2013
DO - 10.1587/transcom.E96.B.522
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E96-B
IS - 2
JA - IEICE TRANSACTIONS on Communications
Y1 - February 2013
AB - We use end-to-end measurements to address the problem of fault diagnosis in computer networks. Since link-level characteristics cannot be uniquely determined from available end-to-end measurements, most existing diagnosis approaches make statistical assumptions of the network to obtain a unique solution. However, the performance of these approaches is not assured due to the uncertainty of the assumptions. Thus the diagnostic accuracy cannot be guaranteed. In this paper, we propose a different paradigm for fault diagnosis which can find all identifiable links and the minimal identifiable link sequences, and infer their loss rates with the least error. Compared with a former representative diagnosis method through experiments, the experimental results show that our method has smaller diagnosis granularity and much less running time for most network topologies. We also conducted experiments using 105 Planetlab hosts. The results validate the performance of our method as well.
ER -