Towards the improvement of performance anomaly prediction

Marat Zhanikeev; Yoshiaki Tanaka

doi:10.1109/CANET.2005.1598205

Towards the improvement of performance anomaly prediction

Marat Zhanikeev^*, Yoshiaki Tanaka

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Growing demand for pro-active abilities in network management requires performance monitoring agents not only to be able to monitor the anomalies, but also to predict future occurrences. Recent research in this area would usually apply a neural network algorithm on raw SNMP or NetFlow data to obtain the knowledge about the patterns in performance data. The results are not always satisfactory due to highly unpredictable nature of cross-traffic in the network. This paper attempts to improve the prediction quality by using data obtained from end-to-end probing. The results prove higher resilience to cross-traffic interference and better pattern recognition.

Original language	English
Title of host publication	First IEEE and IFIP International Conference in Central Asia on Internet, 2005
DOIs	https://doi.org/10.1109/CANET.2005.1598205
Publication status	Published - 2005
Externally published	Yes
Event	First IEEE and IFIP International Conference in Central Asia on Internet, 2005 - Bishkek, Kyrgyzstan Duration: 2005 Sept 26 → 2005 Sept 28

Publication series

Name	First IEEE and IFIP International Conference in Central Asia on Internet, 2005
Volume	2005

Conference

Conference	First IEEE and IFIP International Conference in Central Asia on Internet, 2005
Country/Territory	Kyrgyzstan
City	Bishkek
Period	05/9/26 → 05/9/28

ASJC Scopus subject areas

Engineering(all)

Access to Document

10.1109/CANET.2005.1598205

Cite this

Towards the improvement of performance anomaly prediction. / Zhanikeev, Marat; Tanaka, Yoshiaki.
First IEEE and IFIP International Conference in Central Asia on Internet, 2005. 2005. 1598205 (First IEEE and IFIP International Conference in Central Asia on Internet, 2005; Vol. 2005).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Zhanikeev, M & Tanaka, Y 2005, Towards the improvement of performance anomaly prediction. in First IEEE and IFIP International Conference in Central Asia on Internet, 2005., 1598205, First IEEE and IFIP International Conference in Central Asia on Internet, 2005, vol. 2005, First IEEE and IFIP International Conference in Central Asia on Internet, 2005, Bishkek, Kyrgyzstan, 05/9/26. https://doi.org/10.1109/CANET.2005.1598205

@inproceedings{8dfce4763cdf491eabb667099783935e,

title = "Towards the improvement of performance anomaly prediction",

abstract = "Growing demand for pro-active abilities in network management requires performance monitoring agents not only to be able to monitor the anomalies, but also to predict future occurrences. Recent research in this area would usually apply a neural network algorithm on raw SNMP or NetFlow data to obtain the knowledge about the patterns in performance data. The results are not always satisfactory due to highly unpredictable nature of cross-traffic in the network. This paper attempts to improve the prediction quality by using data obtained from end-to-end probing. The results prove higher resilience to cross-traffic interference and better pattern recognition.",

author = "Marat Zhanikeev and Yoshiaki Tanaka",

year = "2005",

doi = "10.1109/CANET.2005.1598205",

language = "English",

isbn = "0780391799",

series = "First IEEE and IFIP International Conference in Central Asia on Internet, 2005",

booktitle = "First IEEE and IFIP International Conference in Central Asia on Internet, 2005",

note = "First IEEE and IFIP International Conference in Central Asia on Internet, 2005 ; Conference date: 26-09-2005 Through 28-09-2005",

}

TY - GEN

T1 - Towards the improvement of performance anomaly prediction

AU - Zhanikeev, Marat

AU - Tanaka, Yoshiaki

PY - 2005

Y1 - 2005

N2 - Growing demand for pro-active abilities in network management requires performance monitoring agents not only to be able to monitor the anomalies, but also to predict future occurrences. Recent research in this area would usually apply a neural network algorithm on raw SNMP or NetFlow data to obtain the knowledge about the patterns in performance data. The results are not always satisfactory due to highly unpredictable nature of cross-traffic in the network. This paper attempts to improve the prediction quality by using data obtained from end-to-end probing. The results prove higher resilience to cross-traffic interference and better pattern recognition.

AB - Growing demand for pro-active abilities in network management requires performance monitoring agents not only to be able to monitor the anomalies, but also to predict future occurrences. Recent research in this area would usually apply a neural network algorithm on raw SNMP or NetFlow data to obtain the knowledge about the patterns in performance data. The results are not always satisfactory due to highly unpredictable nature of cross-traffic in the network. This paper attempts to improve the prediction quality by using data obtained from end-to-end probing. The results prove higher resilience to cross-traffic interference and better pattern recognition.

UR - http://www.scopus.com/inward/record.url?scp=33847180546&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33847180546&partnerID=8YFLogxK

U2 - 10.1109/CANET.2005.1598205

DO - 10.1109/CANET.2005.1598205

M3 - Conference contribution

AN - SCOPUS:33847180546

SN - 0780391799

SN - 9780780391796

T3 - First IEEE and IFIP International Conference in Central Asia on Internet, 2005

BT - First IEEE and IFIP International Conference in Central Asia on Internet, 2005

T2 - First IEEE and IFIP International Conference in Central Asia on Internet, 2005

Y2 - 26 September 2005 through 28 September 2005

ER -

Towards the improvement of performance anomaly prediction

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this